* [PATCH V4 01/20] xfsprogs: xfs_repair: allow administrators to add older v5 features
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
@ 2021-12-14 8:47 ` Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 02/20] xfsprogs: Move extent count limits to xfs_format.h Chandan Babu R
` (18 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:47 UTC (permalink / raw)
To: linux-xfs; +Cc: Darrick J. Wong, david, Chandan Babu R
From: "Darrick J. Wong" <djwong@kernel.org>
Add to xfs_db the ability to add certain existing features (finobt,
reflink, and rmapbt) to an existing filesystem if it's eligible.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
include/libxfs.h | 1 +
libxfs/libxfs_api_defs.h | 3 +
man/man8/xfs_admin.8 | 30 ++++++
repair/globals.c | 3 +
repair/globals.h | 3 +
repair/phase2.c | 221 +++++++++++++++++++++++++++++++++++++--
repair/rmap.c | 4 +-
repair/xfs_repair.c | 33 ++++++
8 files changed, 286 insertions(+), 12 deletions(-)
diff --git a/include/libxfs.h b/include/libxfs.h
index 24424d0e..9ca1ed86 100644
--- a/include/libxfs.h
+++ b/include/libxfs.h
@@ -77,6 +77,7 @@ struct iomap;
#include "xfs_refcount_btree.h"
#include "xfs_refcount.h"
#include "xfs_btree_staging.h"
+#include "xfs_ag_resv.h"
#ifndef ARRAY_SIZE
#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
diff --git a/libxfs/libxfs_api_defs.h b/libxfs/libxfs_api_defs.h
index b76e6380..c8a4a8a0 100644
--- a/libxfs/libxfs_api_defs.h
+++ b/libxfs/libxfs_api_defs.h
@@ -21,6 +21,8 @@
#define xfs_ag_init_headers libxfs_ag_init_headers
#define xfs_ag_block_count libxfs_ag_block_count
+#define xfs_ag_resv_free libxfs_ag_resv_free
+#define xfs_ag_resv_init libxfs_ag_resv_init
#define xfs_alloc_ag_max_usable libxfs_alloc_ag_max_usable
#define xfs_allocbt_maxrecs libxfs_allocbt_maxrecs
@@ -109,6 +111,7 @@
#define xfs_highbit32 libxfs_highbit32
#define xfs_highbit64 libxfs_highbit64
#define xfs_ialloc_calc_rootino libxfs_ialloc_calc_rootino
+#define xfs_ialloc_read_agi libxfs_ialloc_read_agi
#define xfs_idata_realloc libxfs_idata_realloc
#define xfs_idestroy_fork libxfs_idestroy_fork
#define xfs_iext_lookup_extent libxfs_iext_lookup_extent
diff --git a/man/man8/xfs_admin.8 b/man/man8/xfs_admin.8
index ad28e0f6..4f3c882a 100644
--- a/man/man8/xfs_admin.8
+++ b/man/man8/xfs_admin.8
@@ -149,6 +149,36 @@ Upgrade a filesystem to support larger timestamps up to the year 2486.
The filesystem cannot be downgraded after this feature is enabled.
Once enabled, the filesystem will not be mountable by older kernels.
This feature was added to Linux 5.10.
+.TP 0.4i
+.B finobt
+Track free inodes through a separate free inode btree index to speed up inode
+allocation on old filesystems.
+This upgrade can fail if any AG has less than 1% free space remaining.
+The filesystem cannot be downgraded after this feature is enabled.
+This feature was added to Linux 3.16.
+.TP 0.4i
+.B reflink
+Enable sharing of file data blocks.
+This upgrade can fail if any AG has less than 2% free space remaining.
+The filesystem cannot be downgraded after this feature is enabled.
+This feature was added to Linux 4.9.
+.TP 0.4i
+.B rmapbt
+Store an index of the owners of on-disk blocks.
+This enables much stronger cross-referencing of various metadata structures
+and online repairs to space usage metadata.
+The filesystem cannot be downgraded after this feature is enabled.
+This upgrade can fail if any AG has less than 5% free space remaining.
+This feature was added to Linux 4.8.
+.TP 0.4i
+.B metadir
+Create a directory tree of metadata inodes instead of storing them all in the
+superblock.
+This is required for reverse mapping btrees and reflink support on the realtime
+device.
+The filesystem cannot be downgraded after this feature is enabled.
+This upgrade can fail if any AG has less than 5% free space remaining.
+This feature is not upstream yet.
.RE
.TP
.BI \-U " uuid"
diff --git a/repair/globals.c b/repair/globals.c
index 506a4e72..d89507b1 100644
--- a/repair/globals.c
+++ b/repair/globals.c
@@ -50,6 +50,9 @@ int convert_lazy_count; /* Convert lazy-count mode on/off */
int lazy_count; /* What to set if to if converting */
bool add_inobtcount; /* add inode btree counts to AGI */
bool add_bigtime; /* add support for timestamps up to 2486 */
+bool add_finobt; /* add free inode btrees */
+bool add_reflink; /* add reference count btrees */
+bool add_rmapbt; /* add reverse mapping btrees */
/* misc status variables */
diff --git a/repair/globals.h b/repair/globals.h
index 929b82be..53ff2532 100644
--- a/repair/globals.h
+++ b/repair/globals.h
@@ -91,6 +91,9 @@ extern int convert_lazy_count; /* Convert lazy-count mode on/off */
extern int lazy_count; /* What to set if to if converting */
extern bool add_inobtcount; /* add inode btree counts to AGI */
extern bool add_bigtime; /* add support for timestamps up to 2486 */
+extern bool add_finobt; /* add free inode btrees */
+extern bool add_reflink; /* add reference count btrees */
+extern bool add_rmapbt; /* add reverse mapping btrees */
/* misc status variables */
diff --git a/repair/phase2.c b/repair/phase2.c
index cb9adf1d..c811ed5d 100644
--- a/repair/phase2.c
+++ b/repair/phase2.c
@@ -133,7 +133,8 @@ zero_log(
static bool
set_inobtcount(
- struct xfs_mount *mp)
+ struct xfs_mount *mp,
+ struct xfs_sb *new_sb)
{
if (!xfs_sb_version_hascrc(&mp->m_sb)) {
printf(
@@ -153,14 +154,15 @@ set_inobtcount(
}
printf(_("Adding inode btree counts to filesystem.\n"));
- mp->m_sb.sb_features_ro_compat |= XFS_SB_FEAT_RO_COMPAT_INOBTCNT;
- mp->m_sb.sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
+ new_sb->sb_features_ro_compat |= XFS_SB_FEAT_RO_COMPAT_INOBTCNT;
+ new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
return true;
}
static bool
set_bigtime(
- struct xfs_mount *mp)
+ struct xfs_mount *mp,
+ struct xfs_sb *new_sb)
{
if (!xfs_sb_version_hascrc(&mp->m_sb)) {
printf(
@@ -174,8 +176,193 @@ set_bigtime(
}
printf(_("Adding large timestamp support to filesystem.\n"));
- mp->m_sb.sb_features_incompat |= (XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR |
- XFS_SB_FEAT_INCOMPAT_BIGTIME);
+ new_sb->sb_features_incompat |= (XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR |
+ XFS_SB_FEAT_INCOMPAT_BIGTIME);
+ return true;
+}
+
+/* Make sure we can actually upgrade this (v5) filesystem. */
+static void
+check_new_v5_geometry(
+ struct xfs_mount *mp,
+ struct xfs_sb *new_sb)
+{
+ struct xfs_sb old_sb;
+ struct xfs_perag *pag;
+ xfs_agnumber_t agno;
+ xfs_ino_t rootino;
+ int min_logblocks;
+ int error;
+
+ /*
+ * Save the current superblock, then copy in the new one to do log size
+ * and root inode checks.
+ */
+ memcpy(&old_sb, &mp->m_sb, sizeof(struct xfs_sb));
+ memcpy(&mp->m_sb, new_sb, sizeof(struct xfs_sb));
+
+ /* Do we have a big enough log? */
+ min_logblocks = libxfs_log_calc_minimum_size(mp);
+ if (old_sb.sb_logblocks < min_logblocks) {
+ printf(
+ _("Filesystem log too small to upgrade filesystem; need %u blocks, have %u.\n"),
+ min_logblocks, old_sb.sb_logblocks);
+ exit(0);
+ }
+
+ rootino = libxfs_ialloc_calc_rootino(mp, new_sb->sb_unit);
+ if (old_sb.sb_rootino != rootino) {
+ printf(
+ _("Cannot upgrade filesystem, root inode (%llu) cannot be moved to %llu.\n"),
+ (unsigned long long)old_sb.sb_rootino,
+ (unsigned long long)rootino);
+ exit(0);
+ }
+
+ /* Make sure we have enough space for per-AG reservations. */
+ for_each_perag(mp, agno, pag) {
+ struct xfs_trans *tp;
+ struct xfs_agf *agf;
+ struct xfs_buf *agi_bp, *agf_bp;
+ unsigned int avail, agblocks;
+
+ /*
+ * Create a dummy transaction so that we can load the AGI and
+ * AGF buffers in memory with the old fs geometry and pin them
+ * there while we try to make a per-AG reservation with the new
+ * geometry.
+ */
+ error = -libxfs_trans_alloc_empty(mp, &tp);
+ if (error)
+ do_error(
+ _("Cannot reserve resources for upgrade check, err=%d.\n"),
+ error);
+
+ error = -libxfs_ialloc_read_agi(mp, tp, agno, &agi_bp);
+ if (error)
+ do_error(
+ _("Cannot read AGI %u for upgrade check, err=%d.\n"),
+ agno, error);
+
+ error = -libxfs_alloc_read_agf(mp, tp, agno, 0, &agf_bp);
+ if (error)
+ do_error(
+ _("Cannot read AGF %u for upgrade check, err=%d.\n"),
+ agno, error);
+ agf = agf_bp->b_addr;
+ agblocks = be32_to_cpu(agf->agf_length);
+
+ error = -libxfs_ag_resv_init(pag, tp);
+ if (error == ENOSPC) {
+ printf(
+ _("Not enough free space would remain in AG %u for metadata.\n"),
+ agno);
+ exit(0);
+ }
+ if (error)
+ do_error(
+ _("Error %d while checking AG %u space reservation.\n"),
+ error, agno);
+
+ /*
+ * Would we have at least 10% free space in this AG after
+ * making per-AG reservations?
+ */
+ avail = pag->pagf_freeblks + pag->pagf_flcount;
+ avail -= pag->pag_meta_resv.ar_reserved;
+ avail -= pag->pag_rmapbt_resv.ar_asked;
+ if (avail < agblocks / 10)
+ printf(
+ _("AG %u will be low on space after upgrade.\n"),
+ agno);
+
+ libxfs_ag_resv_free(pag);
+
+ /*
+ * Mark the per-AG structure as uninitialized so that we don't
+ * trip over stale cached counters after the upgrade, and
+ * release all the resources.
+ */
+ libxfs_trans_cancel(tp);
+ pag->pagf_init = 0;
+ pag->pagi_init = 0;
+ }
+
+ /*
+ * Put back the old superblock.
+ */
+ memcpy(&mp->m_sb, &old_sb, sizeof(struct xfs_sb));
+}
+
+static bool
+set_finobt(
+ struct xfs_mount *mp,
+ struct xfs_sb *new_sb)
+{
+ if (!xfs_sb_version_hascrc(&mp->m_sb)) {
+ printf(
+ _("Free inode btree feature only supported on V5 filesystems.\n"));
+ exit(0);
+ }
+
+ if (xfs_sb_version_hasfinobt(&mp->m_sb)) {
+ printf(_("Filesystem already supports free inode btrees.\n"));
+ exit(0);
+ }
+
+ printf(_("Adding free inode btrees to filesystem.\n"));
+ new_sb->sb_features_ro_compat |= XFS_SB_FEAT_RO_COMPAT_FINOBT;
+ new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
+ return true;
+}
+
+static bool
+set_reflink(
+ struct xfs_mount *mp,
+ struct xfs_sb *new_sb)
+{
+ if (!xfs_sb_version_hascrc(&mp->m_sb)) {
+ printf(
+ _("Reflink feature only supported on V5 filesystems.\n"));
+ exit(0);
+ }
+
+ if (xfs_sb_version_hasreflink(&mp->m_sb)) {
+ printf(_("Filesystem already supports reflink.\n"));
+ exit(0);
+ }
+
+ printf(_("Adding reflink support to filesystem.\n"));
+ new_sb->sb_features_ro_compat |= XFS_SB_FEAT_RO_COMPAT_REFLINK;
+ new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
+ return true;
+}
+
+static bool
+set_rmapbt(
+ struct xfs_mount *mp,
+ struct xfs_sb *new_sb)
+{
+ if (!xfs_sb_version_hascrc(&mp->m_sb)) {
+ printf(
+ _("Reverse mapping btree feature only supported on V5 filesystems.\n"));
+ exit(0);
+ }
+
+ if (xfs_sb_version_hasreflink(&mp->m_sb)) {
+ printf(
+ _("Reverse mapping btrees cannot be added when reflink is enabled.\n"));
+ exit(0);
+ }
+
+ if (xfs_sb_version_hasrmapbt(&mp->m_sb)) {
+ printf(_("Filesystem already supports reverse mapping btrees.\n"));
+ exit(0);
+ }
+
+ printf(_("Adding reverse mapping btrees to filesystem.\n"));
+ new_sb->sb_features_ro_compat |= XFS_SB_FEAT_RO_COMPAT_RMAPBT;
+ new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
return true;
}
@@ -184,16 +371,30 @@ static void
upgrade_filesystem(
struct xfs_mount *mp)
{
+ struct xfs_sb new_sb;
struct xfs_buf *bp;
bool dirty = false;
int error;
+ memcpy(&new_sb, &mp->m_sb, sizeof(struct xfs_sb));
+
if (add_inobtcount)
- dirty |= set_inobtcount(mp);
+ dirty |= set_inobtcount(mp, &new_sb);
if (add_bigtime)
- dirty |= set_bigtime(mp);
-
- if (no_modify || !dirty)
+ dirty |= set_bigtime(mp, &new_sb);
+ if (add_finobt)
+ dirty |= set_finobt(mp, &new_sb);
+ if (add_reflink)
+ dirty |= set_reflink(mp, &new_sb);
+ if (add_rmapbt)
+ dirty |= set_rmapbt(mp, &new_sb);
+
+ if (!dirty)
+ return;
+
+ check_new_v5_geometry(mp, &new_sb);
+ memcpy(&mp->m_sb, &new_sb, sizeof(struct xfs_sb));
+ if (no_modify)
return;
bp = libxfs_getsb(mp);
diff --git a/repair/rmap.c b/repair/rmap.c
index 12fe7442..1e8948a6 100644
--- a/repair/rmap.c
+++ b/repair/rmap.c
@@ -49,8 +49,8 @@ bool
rmap_needs_work(
struct xfs_mount *mp)
{
- return xfs_sb_version_hasreflink(&mp->m_sb) ||
- xfs_sb_version_hasrmapbt(&mp->m_sb);
+ return xfs_sb_version_hasreflink(&mp->m_sb) || add_reflink ||
+ xfs_sb_version_hasrmapbt(&mp->m_sb) || add_rmapbt;
}
/*
diff --git a/repair/xfs_repair.c b/repair/xfs_repair.c
index 38406eea..e250a5bf 100644
--- a/repair/xfs_repair.c
+++ b/repair/xfs_repair.c
@@ -67,6 +67,9 @@ enum c_opt_nums {
CONVERT_LAZY_COUNT = 0,
CONVERT_INOBTCOUNT,
CONVERT_BIGTIME,
+ CONVERT_FINOBT,
+ CONVERT_REFLINK,
+ CONVERT_RMAPBT,
C_MAX_OPTS,
};
@@ -74,6 +77,9 @@ static char *c_opts[] = {
[CONVERT_LAZY_COUNT] = "lazycount",
[CONVERT_INOBTCOUNT] = "inobtcount",
[CONVERT_BIGTIME] = "bigtime",
+ [CONVERT_FINOBT] = "finobt",
+ [CONVERT_REFLINK] = "reflink",
+ [CONVERT_RMAPBT] = "rmapbt",
[C_MAX_OPTS] = NULL,
};
@@ -324,6 +330,33 @@ process_args(int argc, char **argv)
_("-c bigtime only supports upgrades\n"));
add_bigtime = true;
break;
+ case CONVERT_FINOBT:
+ if (!val)
+ do_abort(
+ _("-c finobt requires a parameter\n"));
+ if (strtol(val, NULL, 0) != 1)
+ do_abort(
+ _("-c finobt only supports upgrades\n"));
+ add_finobt = true;
+ break;
+ case CONVERT_REFLINK:
+ if (!val)
+ do_abort(
+ _("-c reflink requires a parameter\n"));
+ if (strtol(val, NULL, 0) != 1)
+ do_abort(
+ _("-c reflink only supports upgrades\n"));
+ add_reflink = true;
+ break;
+ case CONVERT_RMAPBT:
+ if (!val)
+ do_abort(
+ _("-c rmapbt requires a parameter\n"));
+ if (strtol(val, NULL, 0) != 1)
+ do_abort(
+ _("-c rmapbt only supports upgrades\n"));
+ add_rmapbt = true;
+ break;
default:
unknown('c', val);
break;
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 02/20] xfsprogs: Move extent count limits to xfs_format.h
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 01/20] xfsprogs: xfs_repair: allow administrators to add older v5 features Chandan Babu R
@ 2021-12-14 8:47 ` Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 03/20] xfsprogs: Introduce xfs_iext_max_nextents() helper Chandan Babu R
` (17 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:47 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
Maximum values associated with extent counters i.e. Maximum extent length,
Maximum data extents and Maximum xattr extents are dictated by the on-disk
format. Hence move these definitions over to xfs_format.h.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_format.h | 7 +++++++
libxfs/xfs_types.h | 7 -------
2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 37570cf0..1e31c31c 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -1051,6 +1051,13 @@ enum xfs_dinode_fmt {
{ XFS_DINODE_FMT_BTREE, "btree" }, \
{ XFS_DINODE_FMT_UUID, "uuid" }
+/*
+ * Max values for extlen, extnum, aextnum.
+ */
+#define MAXEXTLEN ((xfs_extlen_t)0x001fffff) /* 21 bits */
+#define MAXEXTNUM ((xfs_extnum_t)0x7fffffff) /* signed int */
+#define MAXAEXTNUM ((xfs_aextnum_t)0x7fff) /* signed short */
+
/*
* Inode minimum and maximum sizes.
*/
diff --git a/libxfs/xfs_types.h b/libxfs/xfs_types.h
index 0870ef6f..7ff162b1 100644
--- a/libxfs/xfs_types.h
+++ b/libxfs/xfs_types.h
@@ -56,13 +56,6 @@ typedef void * xfs_failaddr_t;
#define NULLFSINO ((xfs_ino_t)-1)
#define NULLAGINO ((xfs_agino_t)-1)
-/*
- * Max values for extlen, extnum, aextnum.
- */
-#define MAXEXTLEN ((xfs_extlen_t)0x001fffff) /* 21 bits */
-#define MAXEXTNUM ((xfs_extnum_t)0x7fffffff) /* signed int */
-#define MAXAEXTNUM ((xfs_aextnum_t)0x7fff) /* signed short */
-
/*
* Minimum and maximum blocksize and sectorsize.
* The blocksize upper limit is pretty much arbitrary.
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 03/20] xfsprogs: Introduce xfs_iext_max_nextents() helper
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 01/20] xfsprogs: xfs_repair: allow administrators to add older v5 features Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 02/20] xfsprogs: Move extent count limits to xfs_format.h Chandan Babu R
@ 2021-12-14 8:47 ` Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 04/20] xfsprogs: Use xfs_extnum_t instead of basic data types Chandan Babu R
` (16 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:47 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
xfs_iext_max_nextents() returns the maximum number of extents possible for one
of data, cow or attribute fork. This helper will be extended further in a
future commit when maximum extent counts associated with data/attribute forks
are increased.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_bmap.c | 9 ++++-----
libxfs/xfs_inode_buf.c | 8 +++-----
libxfs/xfs_inode_fork.c | 2 +-
libxfs/xfs_inode_fork.h | 8 ++++++++
repair/dinode.c | 4 ++--
5 files changed, 18 insertions(+), 13 deletions(-)
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 318d0c06..e7911a2a 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -68,13 +68,12 @@ xfs_bmap_compute_maxlevels(
* ATTR2 we have to assume the worst case scenario of a minimum size
* available.
*/
- if (whichfork == XFS_DATA_FORK) {
- maxleafents = MAXEXTNUM;
+ maxleafents = xfs_iext_max_nextents(whichfork);
+ if (whichfork == XFS_DATA_FORK)
sz = XFS_BMDR_SPACE_CALC(MINDBTPTRS);
- } else {
- maxleafents = MAXAEXTNUM;
+ else
sz = XFS_BMDR_SPACE_CALC(MINABTPTRS);
- }
+
maxrootrecs = xfs_bmdr_maxrecs(sz, 0);
minleafrecs = mp->m_bmap_dmnr[0];
minnoderecs = mp->m_bmap_dmnr[1];
diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
index f98f5c47..46dcfe77 100644
--- a/libxfs/xfs_inode_buf.c
+++ b/libxfs/xfs_inode_buf.c
@@ -334,6 +334,7 @@ xfs_dinode_verify_fork(
int whichfork)
{
uint32_t di_nextents = XFS_DFORK_NEXTENTS(dip, whichfork);
+ xfs_extnum_t max_extents;
switch (XFS_DFORK_FORMAT(dip, whichfork)) {
case XFS_DINODE_FMT_LOCAL:
@@ -355,12 +356,9 @@ xfs_dinode_verify_fork(
return __this_address;
break;
case XFS_DINODE_FMT_BTREE:
- if (whichfork == XFS_ATTR_FORK) {
- if (di_nextents > MAXAEXTNUM)
- return __this_address;
- } else if (di_nextents > MAXEXTNUM) {
+ max_extents = xfs_iext_max_nextents(whichfork);
+ if (di_nextents > max_extents)
return __this_address;
- }
break;
default:
return __this_address;
diff --git a/libxfs/xfs_inode_fork.c b/libxfs/xfs_inode_fork.c
index 1a49c41f..9a7c2c91 100644
--- a/libxfs/xfs_inode_fork.c
+++ b/libxfs/xfs_inode_fork.c
@@ -742,7 +742,7 @@ xfs_iext_count_may_overflow(
if (whichfork == XFS_COW_FORK)
return 0;
- max_exts = (whichfork == XFS_ATTR_FORK) ? MAXAEXTNUM : MAXEXTNUM;
+ max_exts = xfs_iext_max_nextents(whichfork);
if (XFS_TEST_ERROR(false, ip->i_mount, XFS_ERRTAG_REDUCE_MAX_IEXTENTS))
max_exts = 10;
diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
index a6f7897b..370246ee 100644
--- a/libxfs/xfs_inode_fork.h
+++ b/libxfs/xfs_inode_fork.h
@@ -133,6 +133,14 @@ static inline int8_t xfs_ifork_format(struct xfs_ifork *ifp)
return ifp->if_format;
}
+static inline xfs_extnum_t xfs_iext_max_nextents(int whichfork)
+{
+ if (whichfork == XFS_DATA_FORK || whichfork == XFS_COW_FORK)
+ return MAXEXTNUM;
+
+ return MAXAEXTNUM;
+}
+
struct xfs_ifork *xfs_ifork_alloc(enum xfs_dinode_fmt format,
xfs_extnum_t nextents);
struct xfs_ifork *xfs_iext_state_to_fork(struct xfs_inode *ip, int state);
diff --git a/repair/dinode.c b/repair/dinode.c
index f39ab2dc..962285d5 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -1796,7 +1796,7 @@ _("bad nblocks %llu for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
}
}
- if (nextents > MAXEXTNUM) {
+ if (nextents > xfs_iext_max_nextents(XFS_DATA_FORK)) {
do_warn(
_("too many data fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
nextents, lino);
@@ -1819,7 +1819,7 @@ _("bad nextents %d for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
}
}
- if (anextents > MAXAEXTNUM) {
+ if (anextents > xfs_iext_max_nextents(XFS_ATTR_FORK)) {
do_warn(
_("too many attr fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
anextents, lino);
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 04/20] xfsprogs: Use xfs_extnum_t instead of basic data types
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (2 preceding siblings ...)
2021-12-14 8:47 ` [PATCH V4 03/20] xfsprogs: Introduce xfs_iext_max_nextents() helper Chandan Babu R
@ 2021-12-14 8:47 ` Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 05/20] xfsprogs: Introduce xfs_dfork_nextents() helper Chandan Babu R
` (15 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:47 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
xfs_extnum_t is the type to use to declare variables which have values
obtained from xfs_dinode->di_[a]nextents. This commit replaces basic
types (e.g. uint32_t) with xfs_extnum_t for such variables.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
db/bmap.c | 2 +-
db/frag.c | 2 +-
libxfs/xfs_bmap.c | 2 +-
libxfs/xfs_inode_buf.c | 2 +-
libxfs/xfs_inode_fork.c | 2 +-
repair/dinode.c | 2 +-
6 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/db/bmap.c b/db/bmap.c
index fdc70e95..7925a197 100644
--- a/db/bmap.c
+++ b/db/bmap.c
@@ -47,7 +47,7 @@ bmap(
int n;
int nex;
xfs_fsblock_t nextbno;
- int nextents;
+ xfs_extnum_t nextents;
xfs_bmbt_ptr_t *pp;
xfs_bmdr_block_t *rblock;
typnm_t typ;
diff --git a/db/frag.c b/db/frag.c
index cc00672e..ee058aad 100644
--- a/db/frag.c
+++ b/db/frag.c
@@ -273,7 +273,7 @@ process_fork(
int whichfork)
{
extmap_t *extmap;
- int nex;
+ xfs_extnum_t nex;
nex = XFS_DFORK_NEXTENTS(dip, whichfork);
if (!nex)
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index e7911a2a..46e4c1d4 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -48,7 +48,7 @@ xfs_bmap_compute_maxlevels(
{
int level; /* btree level */
uint maxblocks; /* max blocks at this level */
- uint maxleafents; /* max leaf entries possible */
+ xfs_extnum_t maxleafents; /* max leaf entries possible */
int maxrootrecs; /* max records in root block */
int minleafrecs; /* min records in leaf block */
int minnoderecs; /* min records in node block */
diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
index 46dcfe77..2ba84dac 100644
--- a/libxfs/xfs_inode_buf.c
+++ b/libxfs/xfs_inode_buf.c
@@ -333,7 +333,7 @@ xfs_dinode_verify_fork(
struct xfs_mount *mp,
int whichfork)
{
- uint32_t di_nextents = XFS_DFORK_NEXTENTS(dip, whichfork);
+ xfs_extnum_t di_nextents = XFS_DFORK_NEXTENTS(dip, whichfork);
xfs_extnum_t max_extents;
switch (XFS_DFORK_FORMAT(dip, whichfork)) {
diff --git a/libxfs/xfs_inode_fork.c b/libxfs/xfs_inode_fork.c
index 9a7c2c91..12656c74 100644
--- a/libxfs/xfs_inode_fork.c
+++ b/libxfs/xfs_inode_fork.c
@@ -103,7 +103,7 @@ xfs_iformat_extents(
struct xfs_mount *mp = ip->i_mount;
struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
int state = xfs_bmap_fork_to_state(whichfork);
- int nex = XFS_DFORK_NEXTENTS(dip, whichfork);
+ xfs_extnum_t nex = XFS_DFORK_NEXTENTS(dip, whichfork);
int size = nex * sizeof(xfs_bmbt_rec_t);
struct xfs_iext_cursor icur;
struct xfs_bmbt_rec *dp;
diff --git a/repair/dinode.c b/repair/dinode.c
index 962285d5..4d125db4 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -932,7 +932,7 @@ process_exinode(
xfs_bmbt_rec_t *rp;
xfs_fileoff_t first_key;
xfs_fileoff_t last_key;
- int32_t numrecs;
+ xfs_extnum_t numrecs;
int ret;
lino = XFS_AGINO_TO_INO(mp, agno, ino);
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 05/20] xfsprogs: Introduce xfs_dfork_nextents() helper
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (3 preceding siblings ...)
2021-12-14 8:47 ` [PATCH V4 04/20] xfsprogs: Use xfs_extnum_t instead of basic data types Chandan Babu R
@ 2021-12-14 8:47 ` Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 06/20] xfsprogs: Use basic types to define xfs_log_dinode's di_nextents and di_anextents Chandan Babu R
` (14 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:47 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
This commit replaces the macro XFS_DFORK_NEXTENTS() with the helper function
xfs_dfork_nextents(). As of this commit, xfs_dfork_nextents() returns the same
value as XFS_DFORK_NEXTENTS(). A future commit which extends inode's extent
counter fields will add more logic to this helper.
This commit also replaces direct accesses to xfs_dinode->di_[a]nextents
with calls to xfs_dfork_nextents().
No functional changes have been made.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
db/bmap.c | 6 ++---
db/btdump.c | 4 ++--
db/check.c | 28 +++++++++++++---------
db/frag.c | 6 +++--
db/inode.c | 14 ++++++-----
db/metadump.c | 4 ++--
libxfs/xfs_format.h | 4 ----
libxfs/xfs_inode_buf.c | 16 +++++++++----
libxfs/xfs_inode_fork.c | 9 +++----
libxfs/xfs_inode_fork.h | 32 +++++++++++++++++++++++++
repair/attr_repair.c | 2 +-
repair/dinode.c | 53 +++++++++++++++++++++++------------------
repair/prefetch.c | 2 +-
13 files changed, 116 insertions(+), 64 deletions(-)
diff --git a/db/bmap.c b/db/bmap.c
index 7925a197..6b0246d9 100644
--- a/db/bmap.c
+++ b/db/bmap.c
@@ -68,7 +68,7 @@ bmap(
ASSERT(fmt == XFS_DINODE_FMT_LOCAL || fmt == XFS_DINODE_FMT_EXTENTS ||
fmt == XFS_DINODE_FMT_BTREE);
if (fmt == XFS_DINODE_FMT_EXTENTS) {
- nextents = XFS_DFORK_NEXTENTS(dip, whichfork);
+ nextents = xfs_dfork_nextents(dip, whichfork);
xp = (xfs_bmbt_rec_t *)XFS_DFORK_PTR(dip, whichfork);
for (ep = xp; ep < &xp[nextents] && n < nex; ep++) {
if (!bmap_one_extent(ep, &curoffset, eoffset, &n, bep))
@@ -158,9 +158,9 @@ bmap_f(
push_cur();
set_cur_inode(iocur_top->ino);
dip = iocur_top->data;
- if (be32_to_cpu(dip->di_nextents))
+ if (xfs_dfork_data_extents(dip))
dfork = 1;
- if (be16_to_cpu(dip->di_anextents))
+ if (xfs_dfork_attr_extents(dip))
afork = 1;
pop_cur();
}
diff --git a/db/btdump.c b/db/btdump.c
index 920f595b..6d0041b0 100644
--- a/db/btdump.c
+++ b/db/btdump.c
@@ -166,13 +166,13 @@ dump_inode(
dip = iocur_top->data;
if (attrfork) {
- if (!dip->di_anextents ||
+ if (!xfs_dfork_attr_extents(dip) ||
dip->di_aformat != XFS_DINODE_FMT_BTREE) {
dbprintf(_("attr fork not in btree format\n"));
return 0;
}
} else {
- if (!dip->di_nextents ||
+ if (!xfs_dfork_data_extents(dip) ||
dip->di_format != XFS_DINODE_FMT_BTREE) {
dbprintf(_("data fork not in btree format\n"));
return 0;
diff --git a/db/check.c b/db/check.c
index 485e855e..c164fef9 100644
--- a/db/check.c
+++ b/db/check.c
@@ -2713,7 +2713,7 @@ process_exinode(
xfs_bmbt_rec_t *rp;
rp = (xfs_bmbt_rec_t *)XFS_DFORK_PTR(dip, whichfork);
- *nex = XFS_DFORK_NEXTENTS(dip, whichfork);
+ *nex = xfs_dfork_nextents(dip, whichfork);
if (*nex < 0 || *nex > XFS_DFORK_SIZE(dip, mp, whichfork) /
sizeof(xfs_bmbt_rec_t)) {
if (!sflag || id->ilist)
@@ -2737,12 +2737,14 @@ process_inode(
inodata_t *id = NULL;
xfs_ino_t ino;
xfs_extnum_t nextents = 0;
+ xfs_extnum_t dnextents;
int security;
xfs_rfsblock_t totblocks;
xfs_rfsblock_t totdblocks = 0;
xfs_rfsblock_t totiblocks = 0;
dbm_t type;
xfs_extnum_t anextents = 0;
+ xfs_extnum_t danextents;
xfs_rfsblock_t atotdblocks = 0;
xfs_rfsblock_t atotiblocks = 0;
xfs_qcnt_t bc = 0;
@@ -2871,14 +2873,17 @@ process_inode(
error++;
return;
}
+
+ dnextents = xfs_dfork_data_extents(dip);
+ danextents = xfs_dfork_attr_extents(dip);
+
if (verbose || (id && id->ilist) || CHECK_BLIST(bno))
dbprintf(_("inode %lld mode %#o fmt %s "
"afmt %s "
"nex %d anex %d nblk %lld sz %lld%s%s%s%s%s%s%s\n"),
id->ino, mode, fmtnames[(int)dip->di_format],
fmtnames[(int)dip->di_aformat],
- be32_to_cpu(dip->di_nextents),
- be16_to_cpu(dip->di_anextents),
+ dnextents, danextents,
be64_to_cpu(dip->di_nblocks), be64_to_cpu(dip->di_size),
diflags & XFS_DIFLAG_REALTIME ? " rt" : "",
diflags & XFS_DIFLAG_PREALLOC ? " pre" : "",
@@ -2893,25 +2898,26 @@ process_inode(
type = DBM_DIR;
if (dip->di_format == XFS_DINODE_FMT_LOCAL)
break;
- blkmap = blkmap_alloc(be32_to_cpu(dip->di_nextents));
+
+ blkmap = blkmap_alloc(dnextents);
break;
case S_IFREG:
if (diflags & XFS_DIFLAG_REALTIME)
type = DBM_RTDATA;
else if (id->ino == mp->m_sb.sb_rbmino) {
type = DBM_RTBITMAP;
- blkmap = blkmap_alloc(be32_to_cpu(dip->di_nextents));
+ blkmap = blkmap_alloc(dnextents);
addlink_inode(id);
} else if (id->ino == mp->m_sb.sb_rsumino) {
type = DBM_RTSUM;
- blkmap = blkmap_alloc(be32_to_cpu(dip->di_nextents));
+ blkmap = blkmap_alloc(dnextents);
addlink_inode(id);
}
else if (id->ino == mp->m_sb.sb_uquotino ||
id->ino == mp->m_sb.sb_gquotino ||
id->ino == mp->m_sb.sb_pquotino) {
type = DBM_QUOTA;
- blkmap = blkmap_alloc(be32_to_cpu(dip->di_nextents));
+ blkmap = blkmap_alloc(dnextents);
addlink_inode(id);
}
else
@@ -2993,17 +2999,17 @@ process_inode(
be64_to_cpu(dip->di_nblocks), id->ino, totblocks);
error++;
}
- if (nextents != be32_to_cpu(dip->di_nextents)) {
+ if (nextents != dnextents) {
if (v)
dbprintf(_("bad nextents %d for inode %lld, counted %d\n"),
- be32_to_cpu(dip->di_nextents), id->ino, nextents);
+ dnextents, id->ino, nextents);
error++;
}
- if (anextents != be16_to_cpu(dip->di_anextents)) {
+ if (anextents != danextents) {
if (v)
dbprintf(_("bad anextents %d for inode %lld, counted "
"%d\n"),
- be16_to_cpu(dip->di_anextents), id->ino, anextents);
+ danextents, id->ino, anextents);
error++;
}
if (type == DBM_DIR)
diff --git a/db/frag.c b/db/frag.c
index ee058aad..52ff0d8f 100644
--- a/db/frag.c
+++ b/db/frag.c
@@ -262,9 +262,11 @@ process_exinode(
int whichfork)
{
xfs_bmbt_rec_t *rp;
+ xfs_extnum_t nextents;
rp = (xfs_bmbt_rec_t *)XFS_DFORK_PTR(dip, whichfork);
- process_bmbt_reclist(rp, XFS_DFORK_NEXTENTS(dip, whichfork), extmapp);
+ nextents = xfs_dfork_nextents(dip, whichfork);
+ process_bmbt_reclist(rp, nextents, extmapp);
}
static void
@@ -275,7 +277,7 @@ process_fork(
extmap_t *extmap;
xfs_extnum_t nex;
- nex = XFS_DFORK_NEXTENTS(dip, whichfork);
+ nex = xfs_dfork_nextents(dip, whichfork);
if (!nex)
return;
extmap = extmap_alloc(nex);
diff --git a/db/inode.c b/db/inode.c
index 3453c089..9afa6426 100644
--- a/db/inode.c
+++ b/db/inode.c
@@ -275,7 +275,7 @@ inode_a_bmx_count(
return 0;
ASSERT((char *)XFS_DFORK_APTR(dip) - (char *)dip == byteize(startoff));
return dip->di_aformat == XFS_DINODE_FMT_EXTENTS ?
- be16_to_cpu(dip->di_anextents) : 0;
+ xfs_dfork_attr_extents(dip) : 0;
}
static int
@@ -329,6 +329,7 @@ inode_a_size(
{
struct xfs_attr_shortform *asf;
xfs_dinode_t *dip;
+ xfs_extnum_t nextents;
ASSERT(startoff == 0);
ASSERT(idx == 0);
@@ -338,8 +339,8 @@ inode_a_size(
asf = (struct xfs_attr_shortform *)XFS_DFORK_APTR(dip);
return bitize(be16_to_cpu(asf->hdr.totsize));
case XFS_DINODE_FMT_EXTENTS:
- return (int)be16_to_cpu(dip->di_anextents) *
- bitsz(xfs_bmbt_rec_t);
+ nextents = xfs_dfork_attr_extents(dip);
+ return nextents * bitsz(struct xfs_bmbt_rec);
case XFS_DINODE_FMT_BTREE:
return bitize((int)XFS_DFORK_ASIZE(dip, mp));
default:
@@ -500,7 +501,7 @@ inode_u_bmx_count(
dip = obj;
ASSERT((char *)XFS_DFORK_DPTR(dip) - (char *)dip == byteize(startoff));
return dip->di_format == XFS_DINODE_FMT_EXTENTS ?
- be32_to_cpu(dip->di_nextents) : 0;
+ xfs_dfork_data_extents(dip) : 0;
}
static int
@@ -586,6 +587,7 @@ inode_u_size(
int idx)
{
xfs_dinode_t *dip;
+ xfs_extnum_t nextents;
ASSERT(startoff == 0);
ASSERT(idx == 0);
@@ -596,8 +598,8 @@ inode_u_size(
case XFS_DINODE_FMT_LOCAL:
return bitize((int)be64_to_cpu(dip->di_size));
case XFS_DINODE_FMT_EXTENTS:
- return (int)be32_to_cpu(dip->di_nextents) *
- bitsz(xfs_bmbt_rec_t);
+ nextents = xfs_dfork_data_extents(dip);
+ return nextents * bitsz(struct xfs_bmbt_rec);
case XFS_DINODE_FMT_BTREE:
return bitize((int)XFS_DFORK_DSIZE(dip, mp));
case XFS_DINODE_FMT_UUID:
diff --git a/db/metadump.c b/db/metadump.c
index 09d990dc..8a9aec75 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -2282,7 +2282,7 @@ process_exinode(
whichfork = (itype == TYP_ATTR) ? XFS_ATTR_FORK : XFS_DATA_FORK;
- nex = XFS_DFORK_NEXTENTS(dip, whichfork);
+ nex = xfs_dfork_nextents(dip, whichfork);
used = nex * sizeof(xfs_bmbt_rec_t);
if (nex < 0 || used > XFS_DFORK_SIZE(dip, mp, whichfork)) {
if (show_warnings)
@@ -2335,7 +2335,7 @@ static int
process_dev_inode(
xfs_dinode_t *dip)
{
- if (XFS_DFORK_NEXTENTS(dip, XFS_DATA_FORK)) {
+ if (xfs_dfork_data_extents(dip)) {
if (show_warnings)
print_warning("inode %llu has unexpected extents",
(unsigned long long)cur_ino);
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 1e31c31c..68f42fca 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -1107,10 +1107,6 @@ enum xfs_dinode_fmt {
((w) == XFS_DATA_FORK ? \
(dip)->di_format : \
(dip)->di_aformat)
-#define XFS_DFORK_NEXTENTS(dip,w) \
- ((w) == XFS_DATA_FORK ? \
- be32_to_cpu((dip)->di_nextents) : \
- be16_to_cpu((dip)->di_anextents))
/*
* For block and character special files the 32bit dev_t is stored at the
diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
index 2ba84dac..06b6c09f 100644
--- a/libxfs/xfs_inode_buf.c
+++ b/libxfs/xfs_inode_buf.c
@@ -333,9 +333,11 @@ xfs_dinode_verify_fork(
struct xfs_mount *mp,
int whichfork)
{
- xfs_extnum_t di_nextents = XFS_DFORK_NEXTENTS(dip, whichfork);
+ xfs_extnum_t di_nextents;
xfs_extnum_t max_extents;
+ di_nextents = xfs_dfork_nextents(dip, whichfork);
+
switch (XFS_DFORK_FORMAT(dip, whichfork)) {
case XFS_DINODE_FMT_LOCAL:
/*
@@ -402,6 +404,8 @@ xfs_dinode_verify(
uint16_t flags;
uint64_t flags2;
uint64_t di_size;
+ xfs_rfsblock_t nblocks;
+ xfs_extnum_t nextents;
if (dip->di_magic != cpu_to_be16(XFS_DINODE_MAGIC))
return __this_address;
@@ -432,10 +436,12 @@ xfs_dinode_verify(
if ((S_ISLNK(mode) || S_ISDIR(mode)) && di_size == 0)
return __this_address;
+ nextents = xfs_dfork_data_extents(dip);
+ nextents += xfs_dfork_attr_extents(dip);
+ nblocks = be64_to_cpu(dip->di_nblocks);
+
/* Fork checks carried over from xfs_iformat_fork */
- if (mode &&
- be32_to_cpu(dip->di_nextents) + be16_to_cpu(dip->di_anextents) >
- be64_to_cpu(dip->di_nblocks))
+ if (mode && nextents > nblocks)
return __this_address;
if (mode && XFS_DFORK_BOFF(dip) > mp->m_sb.sb_inodesize)
@@ -492,7 +498,7 @@ xfs_dinode_verify(
default:
return __this_address;
}
- if (dip->di_anextents)
+ if (xfs_dfork_attr_extents(dip))
return __this_address;
}
diff --git a/libxfs/xfs_inode_fork.c b/libxfs/xfs_inode_fork.c
index 12656c74..42a7e4e9 100644
--- a/libxfs/xfs_inode_fork.c
+++ b/libxfs/xfs_inode_fork.c
@@ -103,7 +103,7 @@ xfs_iformat_extents(
struct xfs_mount *mp = ip->i_mount;
struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
int state = xfs_bmap_fork_to_state(whichfork);
- xfs_extnum_t nex = XFS_DFORK_NEXTENTS(dip, whichfork);
+ xfs_extnum_t nex = xfs_dfork_nextents(dip, whichfork);
int size = nex * sizeof(xfs_bmbt_rec_t);
struct xfs_iext_cursor icur;
struct xfs_bmbt_rec *dp;
@@ -228,7 +228,7 @@ xfs_iformat_data_fork(
* depend on it.
*/
ip->i_df.if_format = dip->di_format;
- ip->i_df.if_nextents = be32_to_cpu(dip->di_nextents);
+ ip->i_df.if_nextents = xfs_dfork_data_extents(dip);
switch (inode->i_mode & S_IFMT) {
case S_IFIFO:
@@ -293,14 +293,15 @@ xfs_iformat_attr_fork(
struct xfs_inode *ip,
struct xfs_dinode *dip)
{
+ xfs_extnum_t naextents;
int error = 0;
/*
* Initialize the extent count early, as the per-format routines may
* depend on it.
*/
- ip->i_afp = xfs_ifork_alloc(dip->di_aformat,
- be16_to_cpu(dip->di_anextents));
+ naextents = xfs_dfork_attr_extents(dip);
+ ip->i_afp = xfs_ifork_alloc(dip->di_aformat, naextents);
switch (ip->i_afp->if_format) {
case XFS_DINODE_FMT_LOCAL:
diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
index 370246ee..494942ad 100644
--- a/libxfs/xfs_inode_fork.h
+++ b/libxfs/xfs_inode_fork.h
@@ -141,6 +141,38 @@ static inline xfs_extnum_t xfs_iext_max_nextents(int whichfork)
return MAXAEXTNUM;
}
+static inline xfs_extnum_t
+xfs_dfork_data_extents(
+ struct xfs_dinode *dip)
+{
+ return be32_to_cpu(dip->di_nextents);
+}
+
+static inline xfs_extnum_t
+xfs_dfork_attr_extents(
+ struct xfs_dinode *dip)
+{
+ return be16_to_cpu(dip->di_anextents);
+}
+
+static inline xfs_extnum_t
+xfs_dfork_nextents(
+ struct xfs_dinode *dip,
+ int whichfork)
+{
+ switch (whichfork) {
+ case XFS_DATA_FORK:
+ return xfs_dfork_data_extents(dip);
+ case XFS_ATTR_FORK:
+ return xfs_dfork_attr_extents(dip);
+ default:
+ ASSERT(0);
+ break;
+ }
+
+ return 0;
+}
+
struct xfs_ifork *xfs_ifork_alloc(enum xfs_dinode_fmt format,
xfs_extnum_t nextents);
struct xfs_ifork *xfs_iext_state_to_fork(struct xfs_inode *ip, int state);
diff --git a/repair/attr_repair.c b/repair/attr_repair.c
index bc3c2bef..618f4f1e 100644
--- a/repair/attr_repair.c
+++ b/repair/attr_repair.c
@@ -1083,7 +1083,7 @@ process_longform_attr(
bno = blkmap_get(blkmap, 0);
if (bno == NULLFSBLOCK) {
if (dip->di_aformat == XFS_DINODE_FMT_EXTENTS &&
- be16_to_cpu(dip->di_anextents) == 0)
+ xfs_dfork_attr_extents(dip) == 0)
return(0); /* the kernel can handle this state */
do_warn(
_("block 0 of inode %" PRIu64 " attribute fork is missing\n"),
diff --git a/repair/dinode.c b/repair/dinode.c
index 4d125db4..735ba066 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -68,7 +68,7 @@ _("clearing inode %" PRIu64 " attributes\n"), ino_num);
fprintf(stderr,
_("would have cleared inode %" PRIu64 " attributes\n"), ino_num);
- if (be16_to_cpu(dino->di_anextents) != 0) {
+ if (xfs_dfork_attr_extents(dino) != 0) {
if (no_modify)
return(1);
dino->di_anextents = cpu_to_be16(0);
@@ -938,7 +938,7 @@ process_exinode(
lino = XFS_AGINO_TO_INO(mp, agno, ino);
rp = (xfs_bmbt_rec_t *)XFS_DFORK_PTR(dip, whichfork);
*tot = 0;
- numrecs = XFS_DFORK_NEXTENTS(dip, whichfork);
+ numrecs = xfs_dfork_nextents(dip, whichfork);
/*
* We've already decided on the maximum number of extents on the inode,
@@ -1015,7 +1015,7 @@ process_symlink_extlist(xfs_mount_t *mp, xfs_ino_t lino, xfs_dinode_t *dino)
xfs_fileoff_t expected_offset;
xfs_bmbt_rec_t *rp;
xfs_bmbt_irec_t irec;
- int numrecs;
+ xfs_extnum_t numrecs;
int i;
int max_blocks;
@@ -1037,7 +1037,7 @@ _("mismatch between format (%d) and size (%" PRId64 ") in symlink inode %" PRIu6
}
rp = (xfs_bmbt_rec_t *)XFS_DFORK_DPTR(dino);
- numrecs = be32_to_cpu(dino->di_nextents);
+ numrecs = xfs_dfork_data_extents(dino);
/*
* the max # of extents in a symlink inode is equal to the
@@ -1543,6 +1543,8 @@ process_check_sb_inodes(
int *type,
int *dirty)
{
+ xfs_extnum_t nextents;
+
if (lino == mp->m_sb.sb_rootino) {
if (*type != XR_INO_DIR) {
do_warn(_("root inode %" PRIu64 " has bad type 0x%x\n"),
@@ -1597,10 +1599,12 @@ _("realtime summary inode %" PRIu64 " has bad type 0x%x, "),
do_warn(_("would reset to regular file\n"));
}
}
- if (mp->m_sb.sb_rblocks == 0 && dinoc->di_nextents != 0) {
+
+ nextents = xfs_dfork_data_extents(dinoc);
+ if (mp->m_sb.sb_rblocks == 0 && nextents != 0) {
do_warn(
-_("bad # of extents (%u) for realtime summary inode %" PRIu64 "\n"),
- be32_to_cpu(dinoc->di_nextents), lino);
+_("bad # of extents (%d) for realtime summary inode %" PRIu64 "\n"),
+ nextents, lino);
return 1;
}
return 0;
@@ -1618,10 +1622,12 @@ _("realtime bitmap inode %" PRIu64 " has bad type 0x%x, "),
do_warn(_("would reset to regular file\n"));
}
}
- if (mp->m_sb.sb_rblocks == 0 && dinoc->di_nextents != 0) {
+
+ nextents = xfs_dfork_data_extents(dinoc);
+ if (mp->m_sb.sb_rblocks == 0 && nextents != 0) {
do_warn(
-_("bad # of extents (%u) for realtime bitmap inode %" PRIu64 "\n"),
- be32_to_cpu(dinoc->di_nextents), lino);
+_("bad # of extents (%d) for realtime bitmap inode %" PRIu64 "\n"),
+ nextents, lino);
return 1;
}
return 0;
@@ -1780,6 +1786,8 @@ process_inode_blocks_and_extents(
xfs_ino_t lino,
int *dirty)
{
+ xfs_extnum_t dnextents;
+
if (nblocks != be64_to_cpu(dino->di_nblocks)) {
if (!no_modify) {
do_warn(
@@ -1802,20 +1810,19 @@ _("too many data fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
nextents, lino);
return 1;
}
- if (nextents != be32_to_cpu(dino->di_nextents)) {
+
+ dnextents = xfs_dfork_data_extents(dino);
+ if (nextents != dnextents) {
if (!no_modify) {
do_warn(
_("correcting nextents for inode %" PRIu64 ", was %d - counted %" PRIu64 "\n"),
- lino,
- be32_to_cpu(dino->di_nextents),
- nextents);
+ lino, dnextents, nextents);
dino->di_nextents = cpu_to_be32(nextents);
*dirty = 1;
} else {
do_warn(
_("bad nextents %d for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
- be32_to_cpu(dino->di_nextents),
- lino, nextents);
+ dnextents, lino, nextents);
}
}
@@ -1825,19 +1832,19 @@ _("too many attr fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
anextents, lino);
return 1;
}
- if (anextents != be16_to_cpu(dino->di_anextents)) {
+
+ dnextents = xfs_dfork_attr_extents(dino);
+ if (anextents != dnextents) {
if (!no_modify) {
do_warn(
_("correcting anextents for inode %" PRIu64 ", was %d - counted %" PRIu64 "\n"),
- lino,
- be16_to_cpu(dino->di_anextents), anextents);
+ lino, dnextents, anextents);
dino->di_anextents = cpu_to_be16(anextents);
*dirty = 1;
} else {
do_warn(
_("bad anextents %d for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
- be16_to_cpu(dino->di_anextents),
- lino, anextents);
+ dnextents, lino, anextents);
}
}
@@ -1879,7 +1886,7 @@ process_inode_data_fork(
* uses negative values in memory. hence if we see negative numbers
* here, trash it!
*/
- nex = be32_to_cpu(dino->di_nextents);
+ nex = xfs_dfork_data_extents(dino);
if (nex < 0)
*nextents = 1;
else
@@ -2000,7 +2007,7 @@ process_inode_attr_fork(
return 0;
}
- *anextents = be16_to_cpu(dino->di_anextents);
+ *anextents = xfs_dfork_attr_extents(dino);
if (*anextents > be64_to_cpu(dino->di_nblocks))
*anextents = 1;
diff --git a/repair/prefetch.c b/repair/prefetch.c
index 22a0c0c9..fa6a3893 100644
--- a/repair/prefetch.c
+++ b/repair/prefetch.c
@@ -393,7 +393,7 @@ pf_read_exinode(
xfs_dinode_t *dino)
{
pf_read_bmbt_reclist(args, (xfs_bmbt_rec_t *)XFS_DFORK_DPTR(dino),
- be32_to_cpu(dino->di_nextents));
+ xfs_dfork_data_extents(dino));
}
static void
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 06/20] xfsprogs: Use basic types to define xfs_log_dinode's di_nextents and di_anextents
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (4 preceding siblings ...)
2021-12-14 8:47 ` [PATCH V4 05/20] xfsprogs: Introduce xfs_dfork_nextents() helper Chandan Babu R
@ 2021-12-14 8:47 ` Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 07/20] xfsprogs: Promote xfs_extnum_t and xfs_aextnum_t to 64 and 32-bits respectively Chandan Babu R
` (13 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:47 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
A future commit will increase the width of xfs_extnum_t in order to facilitate
larger per-inode extent counters. Hence this patch now uses basic types to
define xfs_log_dinode->[di_nextents|dianextents].
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_log_format.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libxfs/xfs_log_format.h b/libxfs/xfs_log_format.h
index 2c5bcbc1..c13608aa 100644
--- a/libxfs/xfs_log_format.h
+++ b/libxfs/xfs_log_format.h
@@ -396,8 +396,8 @@ struct xfs_log_dinode {
xfs_fsize_t di_size; /* number of bytes in file */
xfs_rfsblock_t di_nblocks; /* # of direct & btree blocks used */
xfs_extlen_t di_extsize; /* basic/minimum extent size for file */
- xfs_extnum_t di_nextents; /* number of extents in data fork */
- xfs_aextnum_t di_anextents; /* number of extents in attribute fork*/
+ uint32_t di_nextents; /* number of extents in data fork */
+ uint16_t di_anextents; /* number of extents in attribute fork*/
uint8_t di_forkoff; /* attr fork offs, <<3 for 64b align */
int8_t di_aformat; /* format of attr fork's data */
uint32_t di_dmevmask; /* DMIG event mask */
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 07/20] xfsprogs: Promote xfs_extnum_t and xfs_aextnum_t to 64 and 32-bits respectively
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (5 preceding siblings ...)
2021-12-14 8:47 ` [PATCH V4 06/20] xfsprogs: Use basic types to define xfs_log_dinode's di_nextents and di_anextents Chandan Babu R
@ 2021-12-14 8:47 ` Chandan Babu R
2021-12-14 8:47 ` [PATCH V4 08/20] xfsprogs: Introduce XFS_SB_FEAT_INCOMPAT_NREXT64 and associated per-fs feature bit Chandan Babu R
` (12 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:47 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
A future commit will introduce a 64-bit on-disk data extent counter and a
32-bit on-disk attr extent counter. This commit promotes xfs_extnum_t and
xfs_aextnum_t to 64 and 32-bits in order to correctly handle in-core versions
of these quantities.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_bmap.c | 4 ++--
libxfs/xfs_inode_fork.c | 2 +-
libxfs/xfs_inode_fork.h | 2 +-
libxfs/xfs_types.h | 4 ++--
repair/dinode.c | 20 ++++++++++----------
repair/dinode.h | 4 ++--
repair/scan.c | 6 +++---
7 files changed, 21 insertions(+), 21 deletions(-)
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 46e4c1d4..04466348 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -46,9 +46,9 @@ xfs_bmap_compute_maxlevels(
xfs_mount_t *mp, /* file system mount structure */
int whichfork) /* data or attr fork */
{
+ xfs_extnum_t maxleafents; /* max leaf entries possible */
int level; /* btree level */
uint maxblocks; /* max blocks at this level */
- xfs_extnum_t maxleafents; /* max leaf entries possible */
int maxrootrecs; /* max records in root block */
int minleafrecs; /* min records in leaf block */
int minnoderecs; /* min records in node block */
@@ -460,7 +460,7 @@ error0:
if (bp_release)
xfs_trans_brelse(NULL, bp);
error_norelse:
- xfs_warn(mp, "%s: BAD after btree leaves for %d extents",
+ xfs_warn(mp, "%s: BAD after btree leaves for %llu extents",
__func__, i);
xfs_err(mp, "%s: CORRUPTED BTREE OR SOMETHING", __func__);
xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
diff --git a/libxfs/xfs_inode_fork.c b/libxfs/xfs_inode_fork.c
index 42a7e4e9..627eb23b 100644
--- a/libxfs/xfs_inode_fork.c
+++ b/libxfs/xfs_inode_fork.c
@@ -115,7 +115,7 @@ xfs_iformat_extents(
* we just bail out rather than crash in kmem_alloc() or memcpy() below.
*/
if (unlikely(size < 0 || size > XFS_DFORK_SIZE(dip, mp, whichfork))) {
- xfs_warn(ip->i_mount, "corrupt inode %Lu ((a)extents = %d).",
+ xfs_warn(ip->i_mount, "corrupt inode %Lu ((a)extents = %llu).",
(unsigned long long) ip->i_ino, nex);
xfs_inode_verifier_error(ip, -EFSCORRUPTED,
"xfs_iformat_extents(1)", dip, sizeof(*dip),
diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
index 494942ad..b34b5c44 100644
--- a/libxfs/xfs_inode_fork.h
+++ b/libxfs/xfs_inode_fork.h
@@ -21,9 +21,9 @@ struct xfs_ifork {
void *if_root; /* extent tree root */
char *if_data; /* inline file data */
} if_u1;
+ xfs_extnum_t if_nextents; /* # of extents in this fork */
short if_broot_bytes; /* bytes allocated for root */
int8_t if_format; /* format of this fork */
- xfs_extnum_t if_nextents; /* # of extents in this fork */
};
/*
diff --git a/libxfs/xfs_types.h b/libxfs/xfs_types.h
index 7ff162b1..89ab3f48 100644
--- a/libxfs/xfs_types.h
+++ b/libxfs/xfs_types.h
@@ -12,8 +12,8 @@ typedef uint32_t xfs_agblock_t; /* blockno in alloc. group */
typedef uint32_t xfs_agino_t; /* inode # within allocation grp */
typedef uint32_t xfs_extlen_t; /* extent length in blocks */
typedef uint32_t xfs_agnumber_t; /* allocation group number */
-typedef int32_t xfs_extnum_t; /* # of extents in a file */
-typedef int16_t xfs_aextnum_t; /* # extents in an attribute fork */
+typedef uint64_t xfs_extnum_t; /* # of extents in a file */
+typedef uint32_t xfs_aextnum_t; /* # extents in an attribute fork */
typedef int64_t xfs_fsize_t; /* bytes in a file */
typedef uint64_t xfs_ufsize_t; /* unsigned bytes in a file */
diff --git a/repair/dinode.c b/repair/dinode.c
index 735ba066..8b6cd60d 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -342,7 +342,7 @@ static int
process_bmbt_reclist_int(
xfs_mount_t *mp,
xfs_bmbt_rec_t *rp,
- int *numrecs,
+ xfs_extnum_t *numrecs,
int type,
xfs_ino_t ino,
xfs_rfsblock_t *tot,
@@ -645,7 +645,7 @@ int
process_bmbt_reclist(
xfs_mount_t *mp,
xfs_bmbt_rec_t *rp,
- int *numrecs,
+ xfs_extnum_t *numrecs,
int type,
xfs_ino_t ino,
xfs_rfsblock_t *tot,
@@ -666,7 +666,7 @@ int
scan_bmbt_reclist(
xfs_mount_t *mp,
xfs_bmbt_rec_t *rp,
- int *numrecs,
+ xfs_extnum_t *numrecs,
int type,
xfs_ino_t ino,
xfs_rfsblock_t *tot,
@@ -1045,7 +1045,7 @@ _("mismatch between format (%d) and size (%" PRId64 ") in symlink inode %" PRIu6
*/
if (numrecs > max_symlink_blocks) {
do_warn(
-_("bad number of extents (%d) in symlink %" PRIu64 " data fork\n"),
+_("bad number of extents (%lu) in symlink %" PRIu64 " data fork\n"),
numrecs, lino);
return(1);
}
@@ -1603,7 +1603,7 @@ _("realtime summary inode %" PRIu64 " has bad type 0x%x, "),
nextents = xfs_dfork_data_extents(dinoc);
if (mp->m_sb.sb_rblocks == 0 && nextents != 0) {
do_warn(
-_("bad # of extents (%d) for realtime summary inode %" PRIu64 "\n"),
+_("bad # of extents (%lu) for realtime summary inode %" PRIu64 "\n"),
nextents, lino);
return 1;
}
@@ -1626,7 +1626,7 @@ _("realtime bitmap inode %" PRIu64 " has bad type 0x%x, "),
nextents = xfs_dfork_data_extents(dinoc);
if (mp->m_sb.sb_rblocks == 0 && nextents != 0) {
do_warn(
-_("bad # of extents (%d) for realtime bitmap inode %" PRIu64 "\n"),
+_("bad # of extents (%lu) for realtime bitmap inode %" PRIu64 "\n"),
nextents, lino);
return 1;
}
@@ -1815,13 +1815,13 @@ _("too many data fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
if (nextents != dnextents) {
if (!no_modify) {
do_warn(
-_("correcting nextents for inode %" PRIu64 ", was %d - counted %" PRIu64 "\n"),
+_("correcting nextents for inode %" PRIu64 ", was %lu - counted %" PRIu64 "\n"),
lino, dnextents, nextents);
dino->di_nextents = cpu_to_be32(nextents);
*dirty = 1;
} else {
do_warn(
-_("bad nextents %d for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
+_("bad nextents %lu for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
dnextents, lino, nextents);
}
}
@@ -1837,13 +1837,13 @@ _("too many attr fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
if (anextents != dnextents) {
if (!no_modify) {
do_warn(
-_("correcting anextents for inode %" PRIu64 ", was %d - counted %" PRIu64 "\n"),
+_("correcting anextents for inode %" PRIu64 ", was %lu - counted %" PRIu64 "\n"),
lino, dnextents, anextents);
dino->di_anextents = cpu_to_be16(anextents);
*dirty = 1;
} else {
do_warn(
-_("bad anextents %d for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
+_("bad anextents %lu for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
dnextents, lino, anextents);
}
}
diff --git a/repair/dinode.h b/repair/dinode.h
index 1bd0e0b7..3aec7bc1 100644
--- a/repair/dinode.h
+++ b/repair/dinode.h
@@ -20,7 +20,7 @@ convert_extent(
int
process_bmbt_reclist(xfs_mount_t *mp,
xfs_bmbt_rec_t *rp,
- int *numrecs,
+ xfs_extnum_t *numrecs,
int type,
xfs_ino_t ino,
xfs_rfsblock_t *tot,
@@ -33,7 +33,7 @@ int
scan_bmbt_reclist(
xfs_mount_t *mp,
xfs_bmbt_rec_t *rp,
- int *numrecs,
+ xfs_extnum_t *numrecs,
int type,
xfs_ino_t ino,
xfs_rfsblock_t *tot,
diff --git a/repair/scan.c b/repair/scan.c
index 4488e084..6b6c0519 100644
--- a/repair/scan.c
+++ b/repair/scan.c
@@ -220,7 +220,7 @@ scan_bmapbt(
xfs_fileoff_t first_key;
xfs_fileoff_t last_key;
char *forkname = get_forkname(whichfork);
- int numrecs;
+ xfs_extnum_t numrecs;
xfs_agnumber_t agno;
xfs_agblock_t agbno;
int state;
@@ -425,7 +425,7 @@ _("couldn't add inode %"PRIu64" bmbt block %"PRIu64" reverse-mapping data."),
if (numrecs > mp->m_bmap_dmxr[0] || (isroot == 0 && numrecs <
mp->m_bmap_dmnr[0])) {
do_warn(
-_("inode %" PRIu64 " bad # of bmap records (%u, min - %u, max - %u)\n"),
+_("inode %" PRIu64 " bad # of bmap records (%lu, min - %u, max - %u)\n"),
ino, numrecs, mp->m_bmap_dmnr[0],
mp->m_bmap_dmxr[0]);
return(1);
@@ -476,7 +476,7 @@ _("out-of-order bmap key (file offset) in inode %" PRIu64 ", %s fork, fsbno %" P
if (numrecs > mp->m_bmap_dmxr[1] || (isroot == 0 && numrecs <
mp->m_bmap_dmnr[1])) {
do_warn(
-_("inode %" PRIu64 " bad # of bmap records (%u, min - %u, max - %u)\n"),
+_("inode %" PRIu64 " bad # of bmap records (%lu, min - %u, max - %u)\n"),
ino, numrecs, mp->m_bmap_dmnr[1], mp->m_bmap_dmxr[1]);
return(1);
}
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 08/20] xfsprogs: Introduce XFS_SB_FEAT_INCOMPAT_NREXT64 and associated per-fs feature bit
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (6 preceding siblings ...)
2021-12-14 8:47 ` [PATCH V4 07/20] xfsprogs: Promote xfs_extnum_t and xfs_aextnum_t to 64 and 32-bits respectively Chandan Babu R
@ 2021-12-14 8:47 ` Chandan Babu R
2021-12-14 8:48 ` [PATCH V4 09/20] xfsprogs: Introduce XFS_FSOP_GEOM_FLAGS_NREXT64 Chandan Babu R
` (11 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:47 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
XFS_SB_FEAT_INCOMPAT_NREXT64 incompat feature bit will be set on filesystems
which support large per-inode extent counters. This commit defines the new
incompat feature bit and the corresponding per-fs feature bit (along with
inline functions to work on it).
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_format.h | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 68f42fca..23ecbc7d 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -469,6 +469,7 @@ xfs_sb_has_ro_compat_feature(
#define XFS_SB_FEAT_INCOMPAT_META_UUID (1 << 2) /* metadata UUID */
#define XFS_SB_FEAT_INCOMPAT_BIGTIME (1 << 3) /* large timestamps */
#define XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR (1 << 4) /* needs xfs_repair */
+#define XFS_SB_FEAT_INCOMPAT_NREXT64 (1 << 5) /* 64-bit data fork extent counter */
#define XFS_SB_FEAT_INCOMPAT_ALL \
(XFS_SB_FEAT_INCOMPAT_FTYPE| \
XFS_SB_FEAT_INCOMPAT_SPINODES| \
@@ -575,6 +576,12 @@ static inline bool xfs_sb_version_hasbigtime(struct xfs_sb *sbp)
(sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_BIGTIME);
}
+static inline bool xfs_sb_version_hasnrext64(struct xfs_sb *sbp)
+{
+ return XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_5 &&
+ (sbp->sb_features_incompat & XFS_SB_FEAT_INCOMPAT_NREXT64);
+}
+
/*
* Inode btree block counter. We record the number of inobt and finobt blocks
* in the AGI header so that we can skip the finobt walk at mount time when
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 09/20] xfsprogs: Introduce XFS_FSOP_GEOM_FLAGS_NREXT64
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (7 preceding siblings ...)
2021-12-14 8:47 ` [PATCH V4 08/20] xfsprogs: Introduce XFS_SB_FEAT_INCOMPAT_NREXT64 and associated per-fs feature bit Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2021-12-14 8:48 ` [PATCH V4 10/20] xfsprogs: Introduce XFS_DIFLAG2_NREXT64 and associated helpers Chandan Babu R
` (10 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
XFS_FSOP_GEOM_FLAGS_NREXT64 indicates that the current filesystem instance
supports 64-bit per-inode extent counters.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_fs.h | 1 +
libxfs/xfs_sb.c | 3 +++
2 files changed, 4 insertions(+)
diff --git a/libxfs/xfs_fs.h b/libxfs/xfs_fs.h
index bde2b4c6..7260b140 100644
--- a/libxfs/xfs_fs.h
+++ b/libxfs/xfs_fs.h
@@ -251,6 +251,7 @@ typedef struct xfs_fsop_resblks {
#define XFS_FSOP_GEOM_FLAGS_REFLINK (1 << 20) /* files can share blocks */
#define XFS_FSOP_GEOM_FLAGS_BIGTIME (1 << 21) /* 64-bit nsec timestamps */
#define XFS_FSOP_GEOM_FLAGS_INOBTCNT (1 << 22) /* inobt btree counter */
+#define XFS_FSOP_GEOM_FLAGS_NREXT64 (1 << 23) /* 64-bit extent counter */
/*
* Minimum and maximum sizes need for growth checks.
diff --git a/libxfs/xfs_sb.c b/libxfs/xfs_sb.c
index b2e214ee..7bea660a 100644
--- a/libxfs/xfs_sb.c
+++ b/libxfs/xfs_sb.c
@@ -1020,6 +1020,9 @@ xfs_fs_geometry(
geo->logsectsize = sbp->sb_logsectsize;
else
geo->logsectsize = BBSIZE;
+ if (xfs_sb_version_hasnrext64(sbp))
+ geo->flags |= XFS_FSOP_GEOM_FLAGS_NREXT64;
+
geo->rtsectsize = sbp->sb_blocksize;
geo->dirblocksize = xfs_dir2_dirblock_bytes(sbp);
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 10/20] xfsprogs: Introduce XFS_DIFLAG2_NREXT64 and associated helpers
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (8 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 09/20] xfsprogs: Introduce XFS_FSOP_GEOM_FLAGS_NREXT64 Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2022-01-05 1:08 ` Darrick J. Wong
2021-12-14 8:48 ` [PATCH V4 11/20] xfsprogs: Use xfs_rfsblock_t to count maximum blocks that can be used by BMBT Chandan Babu R
` (9 subsequent siblings)
19 siblings, 1 reply; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
This commit adds the new per-inode flag XFS_DIFLAG2_NREXT64 to indicate that
an inode supports 64-bit extent counters. This flag is also enabled by default
on newly created inodes when the corresponding filesystem has large extent
counter feature bit (i.e. XFS_FEAT_NREXT64) set.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
db/inode.c | 3 +++
include/xfs_inode.h | 5 +++++
libxfs/xfs_format.h | 10 +++++++++-
libxfs/xfs_ialloc.c | 2 ++
4 files changed, 19 insertions(+), 1 deletion(-)
diff --git a/db/inode.c b/db/inode.c
index 9afa6426..b1f92d36 100644
--- a/db/inode.c
+++ b/db/inode.c
@@ -178,6 +178,9 @@ const field_t inode_v3_flds[] = {
{ "bigtime", FLDT_UINT1,
OI(COFF(flags2) + bitsz(uint64_t) - XFS_DIFLAG2_BIGTIME_BIT - 1), C1,
0, TYP_NONE },
+ { "nrext64", FLDT_UINT1,
+ OI(COFF(flags2) + bitsz(uint64_t) - XFS_DIFLAG2_NREXT64_BIT-1), C1,
+ 0, TYP_NONE },
{ NULL }
};
diff --git a/include/xfs_inode.h b/include/xfs_inode.h
index 08a62d83..79a5c526 100644
--- a/include/xfs_inode.h
+++ b/include/xfs_inode.h
@@ -164,6 +164,11 @@ static inline bool xfs_inode_has_bigtime(struct xfs_inode *ip)
return ip->i_diflags2 & XFS_DIFLAG2_BIGTIME;
}
+static inline bool xfs_inode_has_nrext64(struct xfs_inode *ip)
+{
+ return ip->i_diflags2 & XFS_DIFLAG2_NREXT64;
+}
+
typedef struct cred {
uid_t cr_uid;
gid_t cr_gid;
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 23ecbc7d..58186f2b 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -1180,15 +1180,17 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev)
#define XFS_DIFLAG2_REFLINK_BIT 1 /* file's blocks may be shared */
#define XFS_DIFLAG2_COWEXTSIZE_BIT 2 /* copy on write extent size hint */
#define XFS_DIFLAG2_BIGTIME_BIT 3 /* big timestamps */
+#define XFS_DIFLAG2_NREXT64_BIT 4 /* 64-bit extent counter enabled */
#define XFS_DIFLAG2_DAX (1 << XFS_DIFLAG2_DAX_BIT)
#define XFS_DIFLAG2_REFLINK (1 << XFS_DIFLAG2_REFLINK_BIT)
#define XFS_DIFLAG2_COWEXTSIZE (1 << XFS_DIFLAG2_COWEXTSIZE_BIT)
#define XFS_DIFLAG2_BIGTIME (1 << XFS_DIFLAG2_BIGTIME_BIT)
+#define XFS_DIFLAG2_NREXT64 (1 << XFS_DIFLAG2_NREXT64_BIT)
#define XFS_DIFLAG2_ANY \
(XFS_DIFLAG2_DAX | XFS_DIFLAG2_REFLINK | XFS_DIFLAG2_COWEXTSIZE | \
- XFS_DIFLAG2_BIGTIME)
+ XFS_DIFLAG2_BIGTIME | XFS_DIFLAG2_NREXT64)
static inline bool xfs_dinode_has_bigtime(const struct xfs_dinode *dip)
{
@@ -1196,6 +1198,12 @@ static inline bool xfs_dinode_has_bigtime(const struct xfs_dinode *dip)
(dip->di_flags2 & cpu_to_be64(XFS_DIFLAG2_BIGTIME));
}
+static inline bool xfs_dinode_has_nrext64(const struct xfs_dinode *dip)
+{
+ return dip->di_version >= 3 &&
+ (dip->di_flags2 & cpu_to_be64(XFS_DIFLAG2_NREXT64));
+}
+
/*
* Inode number format:
* low inopblog bits - offset in block
diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
index 570349b8..77501317 100644
--- a/libxfs/xfs_ialloc.c
+++ b/libxfs/xfs_ialloc.c
@@ -2770,6 +2770,8 @@ xfs_ialloc_setup_geometry(
igeo->new_diflags2 = 0;
if (xfs_sb_version_hasbigtime(&mp->m_sb))
igeo->new_diflags2 |= XFS_DIFLAG2_BIGTIME;
+ if (xfs_sb_version_hasnrext64(&mp->m_sb))
+ igeo->new_diflags2 |= XFS_DIFLAG2_NREXT64;
/* Compute inode btree geometry. */
igeo->agino_log = sbp->sb_inopblog + sbp->sb_agblklog;
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH V4 10/20] xfsprogs: Introduce XFS_DIFLAG2_NREXT64 and associated helpers
2021-12-14 8:48 ` [PATCH V4 10/20] xfsprogs: Introduce XFS_DIFLAG2_NREXT64 and associated helpers Chandan Babu R
@ 2022-01-05 1:08 ` Darrick J. Wong
2022-01-05 14:15 ` Chandan Babu R
0 siblings, 1 reply; 34+ messages in thread
From: Darrick J. Wong @ 2022-01-05 1:08 UTC (permalink / raw)
To: Chandan Babu R; +Cc: linux-xfs, david
On Tue, Dec 14, 2021 at 02:18:01PM +0530, Chandan Babu R wrote:
> This commit adds the new per-inode flag XFS_DIFLAG2_NREXT64 to indicate that
> an inode supports 64-bit extent counters. This flag is also enabled by default
> on newly created inodes when the corresponding filesystem has large extent
> counter feature bit (i.e. XFS_FEAT_NREXT64) set.
>
> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
> ---
> db/inode.c | 3 +++
> include/xfs_inode.h | 5 +++++
> libxfs/xfs_format.h | 10 +++++++++-
> libxfs/xfs_ialloc.c | 2 ++
> 4 files changed, 19 insertions(+), 1 deletion(-)
>
> diff --git a/db/inode.c b/db/inode.c
> index 9afa6426..b1f92d36 100644
> --- a/db/inode.c
> +++ b/db/inode.c
> @@ -178,6 +178,9 @@ const field_t inode_v3_flds[] = {
> { "bigtime", FLDT_UINT1,
> OI(COFF(flags2) + bitsz(uint64_t) - XFS_DIFLAG2_BIGTIME_BIT - 1), C1,
> 0, TYP_NONE },
> + { "nrext64", FLDT_UINT1,
> + OI(COFF(flags2) + bitsz(uint64_t) - XFS_DIFLAG2_NREXT64_BIT-1), C1,
Nit: spaces around the '-' operator.
--D
> + 0, TYP_NONE },
> { NULL }
> };
>
> diff --git a/include/xfs_inode.h b/include/xfs_inode.h
> index 08a62d83..79a5c526 100644
> --- a/include/xfs_inode.h
> +++ b/include/xfs_inode.h
> @@ -164,6 +164,11 @@ static inline bool xfs_inode_has_bigtime(struct xfs_inode *ip)
> return ip->i_diflags2 & XFS_DIFLAG2_BIGTIME;
> }
>
> +static inline bool xfs_inode_has_nrext64(struct xfs_inode *ip)
> +{
> + return ip->i_diflags2 & XFS_DIFLAG2_NREXT64;
> +}
> +
> typedef struct cred {
> uid_t cr_uid;
> gid_t cr_gid;
> diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
> index 23ecbc7d..58186f2b 100644
> --- a/libxfs/xfs_format.h
> +++ b/libxfs/xfs_format.h
> @@ -1180,15 +1180,17 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev)
> #define XFS_DIFLAG2_REFLINK_BIT 1 /* file's blocks may be shared */
> #define XFS_DIFLAG2_COWEXTSIZE_BIT 2 /* copy on write extent size hint */
> #define XFS_DIFLAG2_BIGTIME_BIT 3 /* big timestamps */
> +#define XFS_DIFLAG2_NREXT64_BIT 4 /* 64-bit extent counter enabled */
>
> #define XFS_DIFLAG2_DAX (1 << XFS_DIFLAG2_DAX_BIT)
> #define XFS_DIFLAG2_REFLINK (1 << XFS_DIFLAG2_REFLINK_BIT)
> #define XFS_DIFLAG2_COWEXTSIZE (1 << XFS_DIFLAG2_COWEXTSIZE_BIT)
> #define XFS_DIFLAG2_BIGTIME (1 << XFS_DIFLAG2_BIGTIME_BIT)
> +#define XFS_DIFLAG2_NREXT64 (1 << XFS_DIFLAG2_NREXT64_BIT)
>
> #define XFS_DIFLAG2_ANY \
> (XFS_DIFLAG2_DAX | XFS_DIFLAG2_REFLINK | XFS_DIFLAG2_COWEXTSIZE | \
> - XFS_DIFLAG2_BIGTIME)
> + XFS_DIFLAG2_BIGTIME | XFS_DIFLAG2_NREXT64)
>
> static inline bool xfs_dinode_has_bigtime(const struct xfs_dinode *dip)
> {
> @@ -1196,6 +1198,12 @@ static inline bool xfs_dinode_has_bigtime(const struct xfs_dinode *dip)
> (dip->di_flags2 & cpu_to_be64(XFS_DIFLAG2_BIGTIME));
> }
>
> +static inline bool xfs_dinode_has_nrext64(const struct xfs_dinode *dip)
> +{
> + return dip->di_version >= 3 &&
> + (dip->di_flags2 & cpu_to_be64(XFS_DIFLAG2_NREXT64));
> +}
> +
> /*
> * Inode number format:
> * low inopblog bits - offset in block
> diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
> index 570349b8..77501317 100644
> --- a/libxfs/xfs_ialloc.c
> +++ b/libxfs/xfs_ialloc.c
> @@ -2770,6 +2770,8 @@ xfs_ialloc_setup_geometry(
> igeo->new_diflags2 = 0;
> if (xfs_sb_version_hasbigtime(&mp->m_sb))
> igeo->new_diflags2 |= XFS_DIFLAG2_BIGTIME;
> + if (xfs_sb_version_hasnrext64(&mp->m_sb))
> + igeo->new_diflags2 |= XFS_DIFLAG2_NREXT64;
>
> /* Compute inode btree geometry. */
> igeo->agino_log = sbp->sb_inopblog + sbp->sb_agblklog;
> --
> 2.30.2
>
^ permalink raw reply [flat|nested] 34+ messages in thread* Re: [PATCH V4 10/20] xfsprogs: Introduce XFS_DIFLAG2_NREXT64 and associated helpers
2022-01-05 1:08 ` Darrick J. Wong
@ 2022-01-05 14:15 ` Chandan Babu R
0 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2022-01-05 14:15 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: linux-xfs, david
On 05 Jan 2022 at 06:38, Darrick J. Wong wrote:
> On Tue, Dec 14, 2021 at 02:18:01PM +0530, Chandan Babu R wrote:
>> This commit adds the new per-inode flag XFS_DIFLAG2_NREXT64 to indicate that
>> an inode supports 64-bit extent counters. This flag is also enabled by default
>> on newly created inodes when the corresponding filesystem has large extent
>> counter feature bit (i.e. XFS_FEAT_NREXT64) set.
>>
>> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
>> ---
>> db/inode.c | 3 +++
>> include/xfs_inode.h | 5 +++++
>> libxfs/xfs_format.h | 10 +++++++++-
>> libxfs/xfs_ialloc.c | 2 ++
>> 4 files changed, 19 insertions(+), 1 deletion(-)
>>
>> diff --git a/db/inode.c b/db/inode.c
>> index 9afa6426..b1f92d36 100644
>> --- a/db/inode.c
>> +++ b/db/inode.c
>> @@ -178,6 +178,9 @@ const field_t inode_v3_flds[] = {
>> { "bigtime", FLDT_UINT1,
>> OI(COFF(flags2) + bitsz(uint64_t) - XFS_DIFLAG2_BIGTIME_BIT - 1), C1,
>> 0, TYP_NONE },
>> + { "nrext64", FLDT_UINT1,
>> + OI(COFF(flags2) + bitsz(uint64_t) - XFS_DIFLAG2_NREXT64_BIT-1), C1,
>
> Nit: spaces around the '-' operator.
>
Ok. I will fix that.
> --D
>
>> + 0, TYP_NONE },
>> { NULL }
>> };
>>
>> diff --git a/include/xfs_inode.h b/include/xfs_inode.h
>> index 08a62d83..79a5c526 100644
>> --- a/include/xfs_inode.h
>> +++ b/include/xfs_inode.h
>> @@ -164,6 +164,11 @@ static inline bool xfs_inode_has_bigtime(struct xfs_inode *ip)
>> return ip->i_diflags2 & XFS_DIFLAG2_BIGTIME;
>> }
>>
>> +static inline bool xfs_inode_has_nrext64(struct xfs_inode *ip)
>> +{
>> + return ip->i_diflags2 & XFS_DIFLAG2_NREXT64;
>> +}
>> +
>> typedef struct cred {
>> uid_t cr_uid;
>> gid_t cr_gid;
>> diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
>> index 23ecbc7d..58186f2b 100644
>> --- a/libxfs/xfs_format.h
>> +++ b/libxfs/xfs_format.h
>> @@ -1180,15 +1180,17 @@ static inline void xfs_dinode_put_rdev(struct xfs_dinode *dip, xfs_dev_t rdev)
>> #define XFS_DIFLAG2_REFLINK_BIT 1 /* file's blocks may be shared */
>> #define XFS_DIFLAG2_COWEXTSIZE_BIT 2 /* copy on write extent size hint */
>> #define XFS_DIFLAG2_BIGTIME_BIT 3 /* big timestamps */
>> +#define XFS_DIFLAG2_NREXT64_BIT 4 /* 64-bit extent counter enabled */
>>
>> #define XFS_DIFLAG2_DAX (1 << XFS_DIFLAG2_DAX_BIT)
>> #define XFS_DIFLAG2_REFLINK (1 << XFS_DIFLAG2_REFLINK_BIT)
>> #define XFS_DIFLAG2_COWEXTSIZE (1 << XFS_DIFLAG2_COWEXTSIZE_BIT)
>> #define XFS_DIFLAG2_BIGTIME (1 << XFS_DIFLAG2_BIGTIME_BIT)
>> +#define XFS_DIFLAG2_NREXT64 (1 << XFS_DIFLAG2_NREXT64_BIT)
>>
>> #define XFS_DIFLAG2_ANY \
>> (XFS_DIFLAG2_DAX | XFS_DIFLAG2_REFLINK | XFS_DIFLAG2_COWEXTSIZE | \
>> - XFS_DIFLAG2_BIGTIME)
>> + XFS_DIFLAG2_BIGTIME | XFS_DIFLAG2_NREXT64)
>>
>> static inline bool xfs_dinode_has_bigtime(const struct xfs_dinode *dip)
>> {
>> @@ -1196,6 +1198,12 @@ static inline bool xfs_dinode_has_bigtime(const struct xfs_dinode *dip)
>> (dip->di_flags2 & cpu_to_be64(XFS_DIFLAG2_BIGTIME));
>> }
>>
>> +static inline bool xfs_dinode_has_nrext64(const struct xfs_dinode *dip)
>> +{
>> + return dip->di_version >= 3 &&
>> + (dip->di_flags2 & cpu_to_be64(XFS_DIFLAG2_NREXT64));
>> +}
>> +
>> /*
>> * Inode number format:
>> * low inopblog bits - offset in block
>> diff --git a/libxfs/xfs_ialloc.c b/libxfs/xfs_ialloc.c
>> index 570349b8..77501317 100644
>> --- a/libxfs/xfs_ialloc.c
>> +++ b/libxfs/xfs_ialloc.c
>> @@ -2770,6 +2770,8 @@ xfs_ialloc_setup_geometry(
>> igeo->new_diflags2 = 0;
>> if (xfs_sb_version_hasbigtime(&mp->m_sb))
>> igeo->new_diflags2 |= XFS_DIFLAG2_BIGTIME;
>> + if (xfs_sb_version_hasnrext64(&mp->m_sb))
>> + igeo->new_diflags2 |= XFS_DIFLAG2_NREXT64;
>>
>> /* Compute inode btree geometry. */
>> igeo->agino_log = sbp->sb_inopblog + sbp->sb_agblklog;
>> --
>> 2.30.2
>>
--
chandan
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH V4 11/20] xfsprogs: Use xfs_rfsblock_t to count maximum blocks that can be used by BMBT
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (9 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 10/20] xfsprogs: Introduce XFS_DIFLAG2_NREXT64 and associated helpers Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2021-12-14 8:48 ` [PATCH V4 12/20] xfsprogs: Introduce macros to represent new maximum extent counts for data/attr forks Chandan Babu R
` (8 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_bmap.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 04466348..37d9d47c 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -47,8 +47,8 @@ xfs_bmap_compute_maxlevels(
int whichfork) /* data or attr fork */
{
xfs_extnum_t maxleafents; /* max leaf entries possible */
+ xfs_rfsblock_t maxblocks; /* max blocks at this level */
int level; /* btree level */
- uint maxblocks; /* max blocks at this level */
int maxrootrecs; /* max records in root block */
int minleafrecs; /* min records in leaf block */
int minnoderecs; /* min records in node block */
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 12/20] xfsprogs: Introduce macros to represent new maximum extent counts for data/attr forks
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (10 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 11/20] xfsprogs: Use xfs_rfsblock_t to count maximum blocks that can be used by BMBT Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2021-12-14 8:48 ` [PATCH V4 13/20] xfsprogs: Introduce per-inode 64-bit extent counters Chandan Babu R
` (7 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
This commit defines new macros to represent maximum extent counts allowed by
filesystems which have support for large per-inode extent counters.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_bmap.c | 9 ++++-----
libxfs/xfs_format.h | 8 +++++---
libxfs/xfs_inode_buf.c | 3 ++-
libxfs/xfs_inode_fork.c | 2 +-
libxfs/xfs_inode_fork.h | 19 +++++++++++++++----
repair/dinode.c | 6 ++++--
6 files changed, 31 insertions(+), 16 deletions(-)
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 37d9d47c..9dd24678 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -55,10 +55,8 @@ xfs_bmap_compute_maxlevels(
int sz; /* root block size */
/*
- * The maximum number of extents in a file, hence the maximum number of
- * leaf entries, is controlled by the size of the on-disk extent count,
- * either a signed 32-bit number for the data fork, or a signed 16-bit
- * number for the attr fork.
+ * The maximum number of extents in a fork, hence the maximum number of
+ * leaf entries, is controlled by the size of the on-disk extent count.
*
* Note that we can no longer assume that if we are in ATTR1 that the
* fork offset of all the inodes will be
@@ -68,7 +66,8 @@ xfs_bmap_compute_maxlevels(
* ATTR2 we have to assume the worst case scenario of a minimum size
* available.
*/
- maxleafents = xfs_iext_max_nextents(whichfork);
+ maxleafents = xfs_iext_max_nextents(xfs_sb_version_hasnrext64(&mp->m_sb),
+ whichfork);
if (whichfork == XFS_DATA_FORK)
sz = XFS_BMDR_SPACE_CALC(MINDBTPTRS);
else
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 58186f2b..bdd13ec9 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -1061,9 +1061,11 @@ enum xfs_dinode_fmt {
/*
* Max values for extlen, extnum, aextnum.
*/
-#define MAXEXTLEN ((xfs_extlen_t)0x001fffff) /* 21 bits */
-#define MAXEXTNUM ((xfs_extnum_t)0x7fffffff) /* signed int */
-#define MAXAEXTNUM ((xfs_aextnum_t)0x7fff) /* signed short */
+#define MAXEXTLEN ((xfs_extlen_t)0x1fffff) /* 21 bits */
+#define XFS_MAX_EXTCNT_DATA_FORK ((xfs_extnum_t)0xffffffffffff) /* Unsigned 48-bits */
+#define XFS_MAX_EXTCNT_ATTR_FORK ((xfs_aextnum_t)0xffffffff) /* Unsigned 32-bits */
+#define XFS_MAX_EXTCNT_DATA_FORK_OLD ((xfs_extnum_t)0x7fffffff) /* Signed 32-bits */
+#define XFS_MAX_EXTCNT_ATTR_FORK_OLD ((xfs_aextnum_t)0x7fff) /* Signed 16-bits */
/*
* Inode minimum and maximum sizes.
diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
index 06b6c09f..9bddf790 100644
--- a/libxfs/xfs_inode_buf.c
+++ b/libxfs/xfs_inode_buf.c
@@ -358,7 +358,8 @@ xfs_dinode_verify_fork(
return __this_address;
break;
case XFS_DINODE_FMT_BTREE:
- max_extents = xfs_iext_max_nextents(whichfork);
+ max_extents = xfs_iext_max_nextents(xfs_dinode_has_nrext64(dip),
+ whichfork);
if (di_nextents > max_extents)
return __this_address;
break;
diff --git a/libxfs/xfs_inode_fork.c b/libxfs/xfs_inode_fork.c
index 627eb23b..17265401 100644
--- a/libxfs/xfs_inode_fork.c
+++ b/libxfs/xfs_inode_fork.c
@@ -743,7 +743,7 @@ xfs_iext_count_may_overflow(
if (whichfork == XFS_COW_FORK)
return 0;
- max_exts = xfs_iext_max_nextents(whichfork);
+ max_exts = xfs_iext_max_nextents(xfs_inode_has_nrext64(ip), whichfork);
if (XFS_TEST_ERROR(false, ip->i_mount, XFS_ERRTAG_REDUCE_MAX_IEXTENTS))
max_exts = 10;
diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
index b34b5c44..7d5f0015 100644
--- a/libxfs/xfs_inode_fork.h
+++ b/libxfs/xfs_inode_fork.h
@@ -133,12 +133,23 @@ static inline int8_t xfs_ifork_format(struct xfs_ifork *ifp)
return ifp->if_format;
}
-static inline xfs_extnum_t xfs_iext_max_nextents(int whichfork)
+static inline xfs_extnum_t xfs_iext_max_nextents(bool has_big_extcnt,
+ int whichfork)
{
- if (whichfork == XFS_DATA_FORK || whichfork == XFS_COW_FORK)
- return MAXEXTNUM;
+ switch (whichfork) {
+ case XFS_DATA_FORK:
+ case XFS_COW_FORK:
+ return has_big_extcnt ? XFS_MAX_EXTCNT_DATA_FORK
+ : XFS_MAX_EXTCNT_DATA_FORK_OLD;
+
+ case XFS_ATTR_FORK:
+ return has_big_extcnt ? XFS_MAX_EXTCNT_ATTR_FORK
+ : XFS_MAX_EXTCNT_ATTR_FORK_OLD;
- return MAXAEXTNUM;
+ default:
+ ASSERT(0);
+ return 0;
+ }
}
static inline xfs_extnum_t
diff --git a/repair/dinode.c b/repair/dinode.c
index 8b6cd60d..0df84e48 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -1804,7 +1804,8 @@ _("bad nblocks %llu for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
}
}
- if (nextents > xfs_iext_max_nextents(XFS_DATA_FORK)) {
+ if (nextents > xfs_iext_max_nextents(xfs_dinode_has_nrext64(dino),
+ XFS_DATA_FORK)) {
do_warn(
_("too many data fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
nextents, lino);
@@ -1826,7 +1827,8 @@ _("bad nextents %lu for inode %" PRIu64 ", would reset to %" PRIu64 "\n"),
}
}
- if (anextents > xfs_iext_max_nextents(XFS_ATTR_FORK)) {
+ if (anextents > xfs_iext_max_nextents(xfs_dinode_has_nrext64(dino),
+ XFS_ATTR_FORK)) {
do_warn(
_("too many attr fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
anextents, lino);
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 13/20] xfsprogs: Introduce per-inode 64-bit extent counters
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (11 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 12/20] xfsprogs: Introduce macros to represent new maximum extent counts for data/attr forks Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2022-01-05 1:12 ` Darrick J. Wong
2021-12-14 8:48 ` [PATCH V4 14/20] xfsprogs: Conditionally upgrade existing inodes to use " Chandan Babu R
` (6 subsequent siblings)
19 siblings, 1 reply; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david, Dave Chinner
This commit introduces new fields in the on-disk inode format to support
64-bit data fork extent counters and 32-bit attribute fork extent
counters. The new fields will be used only when an inode has
XFS_DIFLAG2_NREXT64 flag set. Otherwise we continue to use the regular 32-bit
data fork extent counters and 16-bit attribute fork extent counters.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Suggested-by: Dave Chinner <dchinner@redhat.com>
---
db/field.c | 4 -
db/field.h | 2 -
db/inode.c | 170 ++++++++++++++++++++++++++++++++++++++-
libxfs/xfs_format.h | 22 ++++-
libxfs/xfs_inode_buf.c | 27 ++++++-
libxfs/xfs_inode_fork.h | 10 ++-
libxfs/xfs_log_format.h | 22 ++++-
logprint/log_misc.c | 20 ++++-
logprint/log_print_all.c | 18 ++++-
repair/dinode.c | 18 ++++-
10 files changed, 279 insertions(+), 34 deletions(-)
diff --git a/db/field.c b/db/field.c
index 51268938..1e274ffc 100644
--- a/db/field.c
+++ b/db/field.c
@@ -25,8 +25,6 @@
#include "symlink.h"
const ftattr_t ftattrtab[] = {
- { FLDT_AEXTNUM, "aextnum", fp_num, "%d", SI(bitsz(xfs_aextnum_t)),
- FTARG_SIGNED, NULL, NULL },
{ FLDT_AGBLOCK, "agblock", fp_num, "%u", SI(bitsz(xfs_agblock_t)),
FTARG_DONULL, fa_agblock, NULL },
{ FLDT_AGBLOCKNZ, "agblocknz", fp_num, "%u", SI(bitsz(xfs_agblock_t)),
@@ -300,8 +298,6 @@ const ftattr_t ftattrtab[] = {
FTARG_DONULL, fa_drtbno, NULL },
{ FLDT_EXTLEN, "extlen", fp_num, "%u", SI(bitsz(xfs_extlen_t)), 0, NULL,
NULL },
- { FLDT_EXTNUM, "extnum", fp_num, "%d", SI(bitsz(xfs_extnum_t)),
- FTARG_SIGNED, NULL, NULL },
{ FLDT_FSIZE, "fsize", fp_num, "%lld", SI(bitsz(xfs_fsize_t)),
FTARG_SIGNED, NULL, NULL },
{ FLDT_INO, "ino", fp_num, "%llu", SI(bitsz(xfs_ino_t)), FTARG_DONULL,
diff --git a/db/field.h b/db/field.h
index 387c189e..614fd0ab 100644
--- a/db/field.h
+++ b/db/field.h
@@ -5,7 +5,6 @@
*/
typedef enum fldt {
- FLDT_AEXTNUM,
FLDT_AGBLOCK,
FLDT_AGBLOCKNZ,
FLDT_AGF,
@@ -143,7 +142,6 @@ typedef enum fldt {
FLDT_DRFSBNO,
FLDT_DRTBNO,
FLDT_EXTLEN,
- FLDT_EXTNUM,
FLDT_FSIZE,
FLDT_INO,
FLDT_INOBT,
diff --git a/db/inode.c b/db/inode.c
index b1f92d36..226797be 100644
--- a/db/inode.c
+++ b/db/inode.c
@@ -27,6 +27,14 @@ static int inode_core_nlinkv2_count(void *obj, int startoff);
static int inode_core_onlink_count(void *obj, int startoff);
static int inode_core_projid_count(void *obj, int startoff);
static int inode_core_nlinkv1_count(void *obj, int startoff);
+static int inode_core_big_dextcnt_count(void *obj, int startoff);
+static int inode_core_v3_pad_count(void *obj, int startoff);
+static int inode_core_v2_pad_count(void *obj, int startoff);
+static int inode_core_flushiter_count(void *obj, int startoff);
+static int inode_core_big_aextcnt_count(void *obj, int startoff);
+static int inode_core_nrext64_pad_count(void *obj, int startoff);
+static int inode_core_nextents_count(void *obj, int startoff);
+static int inode_core_anextents_count(void *obj, int startoff);
static int inode_f(int argc, char **argv);
static int inode_u_offset(void *obj, int startoff, int idx);
static int inode_u_bmbt_count(void *obj, int startoff);
@@ -90,18 +98,30 @@ const field_t inode_core_flds[] = {
inode_core_projid_count, FLD_COUNT, TYP_NONE },
{ "projid_hi", FLDT_UINT16D, OI(COFF(projid_hi)),
inode_core_projid_count, FLD_COUNT, TYP_NONE },
- { "pad", FLDT_UINT8X, OI(OFF(pad)), CI(6), FLD_ARRAY|FLD_SKIPALL, TYP_NONE },
+ { "big_dextcnt", FLDT_UINT64D, OI(COFF(big_dextcnt)),
+ inode_core_big_dextcnt_count, FLD_COUNT, TYP_NONE },
+ { "v3_pad", FLDT_UINT8X, OI(OFF(v3_pad)),
+ inode_core_v3_pad_count, FLD_ARRAY|FLD_COUNT|FLD_SKIPALL, TYP_NONE },
+ { "v2_pad", FLDT_UINT8X, OI(OFF(v2_pad)),
+ inode_core_v2_pad_count, FLD_ARRAY|FLD_COUNT|FLD_SKIPALL, TYP_NONE },
{ "uid", FLDT_UINT32D, OI(COFF(uid)), C1, 0, TYP_NONE },
{ "gid", FLDT_UINT32D, OI(COFF(gid)), C1, 0, TYP_NONE },
- { "flushiter", FLDT_UINT16D, OI(COFF(flushiter)), C1, 0, TYP_NONE },
+ { "flushiter", FLDT_UINT16D, OI(COFF(flushiter)),
+ inode_core_flushiter_count, FLD_COUNT, TYP_NONE },
{ "atime", FLDT_TIMESTAMP, OI(COFF(atime)), C1, 0, TYP_NONE },
{ "mtime", FLDT_TIMESTAMP, OI(COFF(mtime)), C1, 0, TYP_NONE },
{ "ctime", FLDT_TIMESTAMP, OI(COFF(ctime)), C1, 0, TYP_NONE },
{ "size", FLDT_FSIZE, OI(COFF(size)), C1, 0, TYP_NONE },
{ "nblocks", FLDT_DRFSBNO, OI(COFF(nblocks)), C1, 0, TYP_NONE },
{ "extsize", FLDT_EXTLEN, OI(COFF(extsize)), C1, 0, TYP_NONE },
- { "nextents", FLDT_EXTNUM, OI(COFF(nextents)), C1, 0, TYP_NONE },
- { "naextents", FLDT_AEXTNUM, OI(COFF(anextents)), C1, 0, TYP_NONE },
+ { "big_aextcnt", FLDT_UINT32D, OI(COFF(big_aextcnt)),
+ inode_core_big_aextcnt_count, FLD_COUNT, TYP_NONE },
+ { "nrext64_pad", FLDT_UINT16D, OI(COFF(nrext64_pad)),
+ inode_core_nrext64_pad_count, FLD_COUNT, TYP_NONE },
+ { "nextents", FLDT_UINT32D, OI(COFF(nextents)),
+ inode_core_nextents_count, FLD_COUNT, TYP_NONE },
+ { "naextents", FLDT_UINT16D, OI(COFF(anextents)),
+ inode_core_anextents_count, FLD_COUNT, TYP_NONE },
{ "forkoff", FLDT_UINT8D, OI(COFF(forkoff)), C1, 0, TYP_NONE },
{ "aformat", FLDT_DINODE_FMT, OI(COFF(aformat)), C1, 0, TYP_NONE },
{ "dmevmask", FLDT_UINT32X, OI(COFF(dmevmask)), C1, 0, TYP_NONE },
@@ -403,6 +423,148 @@ inode_core_projid_count(
return dic->di_version >= 2;
}
+static int
+inode_core_big_dextcnt_count(
+ void *obj,
+ int startoff)
+{
+ struct xfs_dinode *dic;
+
+ ASSERT(startoff == 0);
+ ASSERT(obj == iocur_top->data);
+ dic = obj;
+
+ if (dic->di_version == 3 &&
+ (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
+ return 1;
+ else
+ return 0;
+}
+
+static int
+inode_core_v3_pad_count(
+ void *obj,
+ int startoff)
+{
+ struct xfs_dinode *dic;
+
+ ASSERT(startoff == 0);
+ ASSERT(obj == iocur_top->data);
+ dic = obj;
+
+ if ((dic->di_version == 3)
+ && !(be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
+ return 8;
+ else
+ return 0;
+}
+
+static int
+inode_core_v2_pad_count(
+ void *obj,
+ int startoff)
+{
+ struct xfs_dinode *dic;
+
+ ASSERT(startoff == 0);
+ ASSERT(obj == iocur_top->data);
+ dic = obj;
+
+ if (dic->di_version == 3)
+ return 0;
+ else
+ return 6;
+}
+
+static int
+inode_core_flushiter_count(
+ void *obj,
+ int startoff)
+{
+ struct xfs_dinode *dic;
+
+ ASSERT(startoff == 0);
+ ASSERT(obj == iocur_top->data);
+ dic = obj;
+
+ if (dic->di_version == 3)
+ return 0;
+ else
+ return 1;
+}
+
+static int
+inode_core_big_aextcnt_count(
+ void *obj,
+ int startoff)
+{
+ struct xfs_dinode *dic;
+
+ ASSERT(startoff == 0);
+ ASSERT(obj == iocur_top->data);
+ dic = obj;
+
+ if (dic->di_version == 3 &&
+ (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
+ return 1;
+ else
+ return 0;
+}
+
+static int
+inode_core_nrext64_pad_count(
+ void *obj,
+ int startoff)
+{
+ struct xfs_dinode *dic;
+
+ ASSERT(startoff == 0);
+ ASSERT(obj == iocur_top->data);
+ dic = obj;
+
+ if (dic->di_version == 3 &&
+ (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
+ return 1;
+ else
+ return 0;
+}
+
+static int
+inode_core_nextents_count(
+ void *obj,
+ int startoff)
+{
+ struct xfs_dinode *dic;
+
+ ASSERT(startoff == 0);
+ ASSERT(obj == iocur_top->data);
+ dic = obj;
+
+ if ((dic->di_version == 3)
+ && (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
+ return 0;
+ else
+ return 1;
+}
+
+static int
+inode_core_anextents_count(
+ void *obj,
+ int startoff)
+{
+ struct xfs_dinode *dic;
+
+ ASSERT(startoff == 0);
+ ASSERT(obj == iocur_top->data);
+ dic = obj;
+
+ if ((dic->di_version == 3)
+ && (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
+ return 0;
+ else
+ return 1;
+}
+
static int
inode_f(
int argc,
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index bdd13ec9..d4c962fc 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -980,16 +980,30 @@ typedef struct xfs_dinode {
__be32 di_nlink; /* number of links to file */
__be16 di_projid_lo; /* lower part of owner's project id */
__be16 di_projid_hi; /* higher part owner's project id */
- __u8 di_pad[6]; /* unused, zeroed space */
- __be16 di_flushiter; /* incremented on flush */
+ union {
+ __be64 di_big_dextcnt; /* NREXT64 data extents */
+ __u8 di_v3_pad[8]; /* !NREXT64 V3 inode zeroed space */
+ struct {
+ __u8 di_v2_pad[6]; /* V2 inode zeroed space */
+ __be16 di_flushiter; /* V2 inode incremented on flush */
+ };
+ };
xfs_timestamp_t di_atime; /* time last accessed */
xfs_timestamp_t di_mtime; /* time last modified */
xfs_timestamp_t di_ctime; /* time created/inode modified */
__be64 di_size; /* number of bytes in file */
__be64 di_nblocks; /* # of direct & btree blocks used */
__be32 di_extsize; /* basic/minimum extent size for file */
- __be32 di_nextents; /* number of extents in data fork */
- __be16 di_anextents; /* number of extents in attribute fork*/
+ union {
+ struct {
+ __be32 di_big_aextcnt; /* NREXT64 attr extents */
+ __be16 di_nrext64_pad; /* NREXT64 unused, zero */
+ } __packed;
+ struct {
+ __be32 di_nextents; /* !NREXT64 data extents */
+ __be16 di_anextents; /* !NREXT64 attr extents */
+ } __packed;
+ };
__u8 di_forkoff; /* attr fork offs, <<3 for 64b align */
__s8 di_aformat; /* format of attr fork's data */
__be32 di_dmevmask; /* DMIG event mask */
diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
index 9bddf790..5c6de73b 100644
--- a/libxfs/xfs_inode_buf.c
+++ b/libxfs/xfs_inode_buf.c
@@ -276,6 +276,25 @@ xfs_inode_to_disk_ts(
return ts;
}
+static inline void
+xfs_inode_to_disk_iext_counters(
+ struct xfs_inode *ip,
+ struct xfs_dinode *to)
+{
+ if (xfs_inode_has_nrext64(ip)) {
+ to->di_big_dextcnt = cpu_to_be64(xfs_ifork_nextents(&ip->i_df));
+ to->di_big_aextcnt = cpu_to_be32(xfs_ifork_nextents(ip->i_afp));
+ /*
+ * We might be upgrading the inode to use larger extent counters
+ * than was previously used. Hence zero the unused field.
+ */
+ to->di_nrext64_pad = cpu_to_be16(0);
+ } else {
+ to->di_nextents = cpu_to_be32(xfs_ifork_nextents(&ip->i_df));
+ to->di_anextents = cpu_to_be16(xfs_ifork_nextents(ip->i_afp));
+ }
+}
+
void
xfs_inode_to_disk(
struct xfs_inode *ip,
@@ -293,7 +312,6 @@ xfs_inode_to_disk(
to->di_projid_lo = cpu_to_be16(ip->i_projid & 0xffff);
to->di_projid_hi = cpu_to_be16(ip->i_projid >> 16);
- memset(to->di_pad, 0, sizeof(to->di_pad));
to->di_atime = xfs_inode_to_disk_ts(ip, inode->i_atime);
to->di_mtime = xfs_inode_to_disk_ts(ip, inode->i_mtime);
to->di_ctime = xfs_inode_to_disk_ts(ip, inode->i_ctime);
@@ -304,8 +322,6 @@ xfs_inode_to_disk(
to->di_size = cpu_to_be64(ip->i_disk_size);
to->di_nblocks = cpu_to_be64(ip->i_nblocks);
to->di_extsize = cpu_to_be32(ip->i_extsize);
- to->di_nextents = cpu_to_be32(xfs_ifork_nextents(&ip->i_df));
- to->di_anextents = cpu_to_be16(xfs_ifork_nextents(ip->i_afp));
to->di_forkoff = ip->i_forkoff;
to->di_aformat = xfs_ifork_format(ip->i_afp);
to->di_flags = cpu_to_be16(ip->i_diflags);
@@ -320,11 +336,14 @@ xfs_inode_to_disk(
to->di_lsn = cpu_to_be64(lsn);
memset(to->di_pad2, 0, sizeof(to->di_pad2));
uuid_copy(&to->di_uuid, &ip->i_mount->m_sb.sb_meta_uuid);
- to->di_flushiter = 0;
+ memset(to->di_v3_pad, 0, sizeof(to->di_v3_pad));
} else {
to->di_version = 2;
to->di_flushiter = cpu_to_be16(ip->i_flushiter);
+ memset(to->di_v2_pad, 0, sizeof(to->di_v2_pad));
}
+
+ xfs_inode_to_disk_iext_counters(ip, to);
}
static xfs_failaddr_t
diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
index 7d5f0015..03036a2c 100644
--- a/libxfs/xfs_inode_fork.h
+++ b/libxfs/xfs_inode_fork.h
@@ -156,14 +156,20 @@ static inline xfs_extnum_t
xfs_dfork_data_extents(
struct xfs_dinode *dip)
{
- return be32_to_cpu(dip->di_nextents);
+ if (xfs_dinode_has_nrext64(dip))
+ return be64_to_cpu(dip->di_big_dextcnt);
+ else
+ return be32_to_cpu(dip->di_nextents);
}
static inline xfs_extnum_t
xfs_dfork_attr_extents(
struct xfs_dinode *dip)
{
- return be16_to_cpu(dip->di_anextents);
+ if (xfs_dinode_has_nrext64(dip))
+ return be32_to_cpu(dip->di_big_aextcnt);
+ else
+ return be16_to_cpu(dip->di_anextents);
}
static inline xfs_extnum_t
diff --git a/libxfs/xfs_log_format.h b/libxfs/xfs_log_format.h
index c13608aa..8f8c9869 100644
--- a/libxfs/xfs_log_format.h
+++ b/libxfs/xfs_log_format.h
@@ -388,16 +388,30 @@ struct xfs_log_dinode {
uint32_t di_nlink; /* number of links to file */
uint16_t di_projid_lo; /* lower part of owner's project id */
uint16_t di_projid_hi; /* higher part of owner's project id */
- uint8_t di_pad[6]; /* unused, zeroed space */
- uint16_t di_flushiter; /* incremented on flush */
+ union {
+ uint64_t di_big_dextcnt; /* NREXT64 data extents */
+ uint8_t di_v3_pad[8]; /* !NREXT64 V3 inode zeroed space */
+ struct {
+ uint8_t di_v2_pad[6]; /* V2 inode zeroed space */
+ uint16_t di_flushiter; /* V2 inode incremented on flush */
+ };
+ };
xfs_log_timestamp_t di_atime; /* time last accessed */
xfs_log_timestamp_t di_mtime; /* time last modified */
xfs_log_timestamp_t di_ctime; /* time created/inode modified */
xfs_fsize_t di_size; /* number of bytes in file */
xfs_rfsblock_t di_nblocks; /* # of direct & btree blocks used */
xfs_extlen_t di_extsize; /* basic/minimum extent size for file */
- uint32_t di_nextents; /* number of extents in data fork */
- uint16_t di_anextents; /* number of extents in attribute fork*/
+ union {
+ struct {
+ uint32_t di_big_aextcnt; /* NREXT64 attr extents */
+ uint16_t di_nrext64_pad; /* NREXT64 unused, zero */
+ } __packed;
+ struct {
+ uint32_t di_nextents; /* !NREXT64 data extents */
+ uint16_t di_anextents; /* !NREXT64 attr extents */
+ } __packed;
+ };
uint8_t di_forkoff; /* attr fork offs, <<3 for 64b align */
int8_t di_aformat; /* format of attr fork's data */
uint32_t di_dmevmask; /* DMIG event mask */
diff --git a/logprint/log_misc.c b/logprint/log_misc.c
index 35e926a3..721c2eb5 100644
--- a/logprint/log_misc.c
+++ b/logprint/log_misc.c
@@ -440,6 +440,8 @@ static void
xlog_print_trans_inode_core(
struct xfs_log_dinode *ip)
{
+ xfs_extnum_t nextents;
+
printf(_("INODE CORE\n"));
printf(_("magic 0x%hx mode 0%ho version %d format %d\n"),
ip->di_magic, ip->di_mode, (int)ip->di_version,
@@ -450,11 +452,21 @@ xlog_print_trans_inode_core(
xlog_extract_dinode_ts(ip->di_atime),
xlog_extract_dinode_ts(ip->di_mtime),
xlog_extract_dinode_ts(ip->di_ctime));
- printf(_("size 0x%llx nblocks 0x%llx extsize 0x%x nextents 0x%x\n"),
+
+ if (ip->di_flags2 & XFS_DIFLAG2_NREXT64)
+ nextents = ip->di_big_dextcnt;
+ else
+ nextents = ip->di_nextents;
+ printf(_("size 0x%llx nblocks 0x%llx extsize 0x%x nextents 0x%lx\n"),
(unsigned long long)ip->di_size, (unsigned long long)ip->di_nblocks,
- ip->di_extsize, ip->di_nextents);
- printf(_("naextents 0x%x forkoff %d dmevmask 0x%x dmstate 0x%hx\n"),
- ip->di_anextents, (int)ip->di_forkoff, ip->di_dmevmask,
+ ip->di_extsize, nextents);
+
+ if (ip->di_flags2 & XFS_DIFLAG2_NREXT64)
+ nextents = ip->di_big_aextcnt;
+ else
+ nextents = ip->di_anextents;
+ printf(_("naextents 0x%lx forkoff %d dmevmask 0x%x dmstate 0x%hx\n"),
+ nextents, (int)ip->di_forkoff, ip->di_dmevmask,
ip->di_dmstate);
printf(_("flags 0x%x gen 0x%x\n"),
ip->di_flags, ip->di_gen);
diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
index c9c453f6..5e387f38 100644
--- a/logprint/log_print_all.c
+++ b/logprint/log_print_all.c
@@ -240,7 +240,10 @@ STATIC void
xlog_recover_print_inode_core(
struct xfs_log_dinode *di)
{
- printf(_(" CORE inode:\n"));
+ xfs_extnum_t nextents;
+ xfs_aextnum_t anextents;
+
+ printf(_(" CORE inode:\n"));
if (!print_inode)
return;
printf(_(" magic:%c%c mode:0x%x ver:%d format:%d\n"),
@@ -254,10 +257,19 @@ xlog_recover_print_inode_core(
xlog_extract_dinode_ts(di->di_mtime),
xlog_extract_dinode_ts(di->di_ctime));
printf(_(" flushiter:%d\n"), di->di_flushiter);
+
+ if (di->di_flags2 & XFS_DIFLAG2_NREXT64) {
+ nextents = di->di_big_dextcnt;
+ anextents = di->di_big_aextcnt;
+ } else {
+ nextents = di->di_nextents;
+ anextents = di->di_anextents;
+ }
+
printf(_(" size:0x%llx nblks:0x%llx exsize:%d "
- "nextents:%d anextents:%d\n"), (unsigned long long)
+ "nextents:%lu anextents:%u\n"), (unsigned long long)
di->di_size, (unsigned long long)di->di_nblocks,
- di->di_extsize, di->di_nextents, (int)di->di_anextents);
+ di->di_extsize, nextents, anextents);
printf(_(" forkoff:%d dmevmask:0x%x dmstate:%d flags:0x%x "
"gen:%u\n"),
(int)di->di_forkoff, di->di_dmevmask, (int)di->di_dmstate,
diff --git a/repair/dinode.c b/repair/dinode.c
index 0df84e48..3be2e1d5 100644
--- a/repair/dinode.c
+++ b/repair/dinode.c
@@ -71,7 +71,12 @@ _("would have cleared inode %" PRIu64 " attributes\n"), ino_num);
if (xfs_dfork_attr_extents(dino) != 0) {
if (no_modify)
return(1);
- dino->di_anextents = cpu_to_be16(0);
+
+ if (xfs_dinode_has_nrext64(dino))
+ dino->di_big_aextcnt = 0;
+ else
+ dino->di_anextents = 0;
+
}
if (dino->di_aformat != XFS_DINODE_FMT_EXTENTS) {
@@ -1818,7 +1823,10 @@ _("too many data fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
do_warn(
_("correcting nextents for inode %" PRIu64 ", was %lu - counted %" PRIu64 "\n"),
lino, dnextents, nextents);
- dino->di_nextents = cpu_to_be32(nextents);
+ if (xfs_dinode_has_nrext64(dino))
+ dino->di_big_dextcnt = cpu_to_be64(nextents);
+ else
+ dino->di_nextents = cpu_to_be32(nextents);
*dirty = 1;
} else {
do_warn(
@@ -1841,7 +1849,11 @@ _("too many attr fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
do_warn(
_("correcting anextents for inode %" PRIu64 ", was %lu - counted %" PRIu64 "\n"),
lino, dnextents, anextents);
- dino->di_anextents = cpu_to_be16(anextents);
+ if (xfs_dinode_has_nrext64(dino))
+ dino->di_big_aextcnt = cpu_to_be32(anextents);
+ else
+ dino->di_anextents = cpu_to_be16(anextents);
+
*dirty = 1;
} else {
do_warn(
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH V4 13/20] xfsprogs: Introduce per-inode 64-bit extent counters
2021-12-14 8:48 ` [PATCH V4 13/20] xfsprogs: Introduce per-inode 64-bit extent counters Chandan Babu R
@ 2022-01-05 1:12 ` Darrick J. Wong
2022-01-07 16:10 ` Chandan Babu R
0 siblings, 1 reply; 34+ messages in thread
From: Darrick J. Wong @ 2022-01-05 1:12 UTC (permalink / raw)
To: Chandan Babu R; +Cc: linux-xfs, david, Dave Chinner
On Tue, Dec 14, 2021 at 02:18:04PM +0530, Chandan Babu R wrote:
> This commit introduces new fields in the on-disk inode format to support
> 64-bit data fork extent counters and 32-bit attribute fork extent
> counters. The new fields will be used only when an inode has
> XFS_DIFLAG2_NREXT64 flag set. Otherwise we continue to use the regular 32-bit
> data fork extent counters and 16-bit attribute fork extent counters.
>
> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
> Suggested-by: Dave Chinner <dchinner@redhat.com>
> ---
> db/field.c | 4 -
> db/field.h | 2 -
> db/inode.c | 170 ++++++++++++++++++++++++++++++++++++++-
> libxfs/xfs_format.h | 22 ++++-
> libxfs/xfs_inode_buf.c | 27 ++++++-
> libxfs/xfs_inode_fork.h | 10 ++-
> libxfs/xfs_log_format.h | 22 ++++-
> logprint/log_misc.c | 20 ++++-
> logprint/log_print_all.c | 18 ++++-
> repair/dinode.c | 18 ++++-
> 10 files changed, 279 insertions(+), 34 deletions(-)
>
> diff --git a/db/field.c b/db/field.c
> index 51268938..1e274ffc 100644
> --- a/db/field.c
> +++ b/db/field.c
> @@ -25,8 +25,6 @@
> #include "symlink.h"
>
> const ftattr_t ftattrtab[] = {
> - { FLDT_AEXTNUM, "aextnum", fp_num, "%d", SI(bitsz(xfs_aextnum_t)),
> - FTARG_SIGNED, NULL, NULL },
> { FLDT_AGBLOCK, "agblock", fp_num, "%u", SI(bitsz(xfs_agblock_t)),
> FTARG_DONULL, fa_agblock, NULL },
> { FLDT_AGBLOCKNZ, "agblocknz", fp_num, "%u", SI(bitsz(xfs_agblock_t)),
> @@ -300,8 +298,6 @@ const ftattr_t ftattrtab[] = {
> FTARG_DONULL, fa_drtbno, NULL },
> { FLDT_EXTLEN, "extlen", fp_num, "%u", SI(bitsz(xfs_extlen_t)), 0, NULL,
> NULL },
> - { FLDT_EXTNUM, "extnum", fp_num, "%d", SI(bitsz(xfs_extnum_t)),
> - FTARG_SIGNED, NULL, NULL },
> { FLDT_FSIZE, "fsize", fp_num, "%lld", SI(bitsz(xfs_fsize_t)),
> FTARG_SIGNED, NULL, NULL },
> { FLDT_INO, "ino", fp_num, "%llu", SI(bitsz(xfs_ino_t)), FTARG_DONULL,
> diff --git a/db/field.h b/db/field.h
> index 387c189e..614fd0ab 100644
> --- a/db/field.h
> +++ b/db/field.h
> @@ -5,7 +5,6 @@
> */
>
> typedef enum fldt {
> - FLDT_AEXTNUM,
> FLDT_AGBLOCK,
> FLDT_AGBLOCKNZ,
> FLDT_AGF,
> @@ -143,7 +142,6 @@ typedef enum fldt {
> FLDT_DRFSBNO,
> FLDT_DRTBNO,
> FLDT_EXTLEN,
> - FLDT_EXTNUM,
> FLDT_FSIZE,
> FLDT_INO,
> FLDT_INOBT,
> diff --git a/db/inode.c b/db/inode.c
> index b1f92d36..226797be 100644
> --- a/db/inode.c
> +++ b/db/inode.c
> @@ -27,6 +27,14 @@ static int inode_core_nlinkv2_count(void *obj, int startoff);
> static int inode_core_onlink_count(void *obj, int startoff);
> static int inode_core_projid_count(void *obj, int startoff);
> static int inode_core_nlinkv1_count(void *obj, int startoff);
> +static int inode_core_big_dextcnt_count(void *obj, int startoff);
> +static int inode_core_v3_pad_count(void *obj, int startoff);
> +static int inode_core_v2_pad_count(void *obj, int startoff);
> +static int inode_core_flushiter_count(void *obj, int startoff);
> +static int inode_core_big_aextcnt_count(void *obj, int startoff);
> +static int inode_core_nrext64_pad_count(void *obj, int startoff);
> +static int inode_core_nextents_count(void *obj, int startoff);
> +static int inode_core_anextents_count(void *obj, int startoff);
> static int inode_f(int argc, char **argv);
> static int inode_u_offset(void *obj, int startoff, int idx);
> static int inode_u_bmbt_count(void *obj, int startoff);
> @@ -90,18 +98,30 @@ const field_t inode_core_flds[] = {
> inode_core_projid_count, FLD_COUNT, TYP_NONE },
> { "projid_hi", FLDT_UINT16D, OI(COFF(projid_hi)),
> inode_core_projid_count, FLD_COUNT, TYP_NONE },
> - { "pad", FLDT_UINT8X, OI(OFF(pad)), CI(6), FLD_ARRAY|FLD_SKIPALL, TYP_NONE },
> + { "big_dextcnt", FLDT_UINT64D, OI(COFF(big_dextcnt)),
Any reason we don't leave the display names as nextents and naextents?
I think fstests has a fair amount of hardcoded xfs_db output, right?
> + inode_core_big_dextcnt_count, FLD_COUNT, TYP_NONE },
> + { "v3_pad", FLDT_UINT8X, OI(OFF(v3_pad)),
> + inode_core_v3_pad_count, FLD_ARRAY|FLD_COUNT|FLD_SKIPALL, TYP_NONE },
> + { "v2_pad", FLDT_UINT8X, OI(OFF(v2_pad)),
> + inode_core_v2_pad_count, FLD_ARRAY|FLD_COUNT|FLD_SKIPALL, TYP_NONE },
> { "uid", FLDT_UINT32D, OI(COFF(uid)), C1, 0, TYP_NONE },
> { "gid", FLDT_UINT32D, OI(COFF(gid)), C1, 0, TYP_NONE },
> - { "flushiter", FLDT_UINT16D, OI(COFF(flushiter)), C1, 0, TYP_NONE },
> + { "flushiter", FLDT_UINT16D, OI(COFF(flushiter)),
> + inode_core_flushiter_count, FLD_COUNT, TYP_NONE },
> { "atime", FLDT_TIMESTAMP, OI(COFF(atime)), C1, 0, TYP_NONE },
> { "mtime", FLDT_TIMESTAMP, OI(COFF(mtime)), C1, 0, TYP_NONE },
> { "ctime", FLDT_TIMESTAMP, OI(COFF(ctime)), C1, 0, TYP_NONE },
> { "size", FLDT_FSIZE, OI(COFF(size)), C1, 0, TYP_NONE },
> { "nblocks", FLDT_DRFSBNO, OI(COFF(nblocks)), C1, 0, TYP_NONE },
> { "extsize", FLDT_EXTLEN, OI(COFF(extsize)), C1, 0, TYP_NONE },
> - { "nextents", FLDT_EXTNUM, OI(COFF(nextents)), C1, 0, TYP_NONE },
> - { "naextents", FLDT_AEXTNUM, OI(COFF(anextents)), C1, 0, TYP_NONE },
> + { "big_aextcnt", FLDT_UINT32D, OI(COFF(big_aextcnt)),
> + inode_core_big_aextcnt_count, FLD_COUNT, TYP_NONE },
> + { "nrext64_pad", FLDT_UINT16D, OI(COFF(nrext64_pad)),
> + inode_core_nrext64_pad_count, FLD_COUNT, TYP_NONE },
> + { "nextents", FLDT_UINT32D, OI(COFF(nextents)),
> + inode_core_nextents_count, FLD_COUNT, TYP_NONE },
> + { "naextents", FLDT_UINT16D, OI(COFF(anextents)),
> + inode_core_anextents_count, FLD_COUNT, TYP_NONE },
> { "forkoff", FLDT_UINT8D, OI(COFF(forkoff)), C1, 0, TYP_NONE },
> { "aformat", FLDT_DINODE_FMT, OI(COFF(aformat)), C1, 0, TYP_NONE },
> { "dmevmask", FLDT_UINT32X, OI(COFF(dmevmask)), C1, 0, TYP_NONE },
> @@ -403,6 +423,148 @@ inode_core_projid_count(
> return dic->di_version >= 2;
> }
>
> +static int
> +inode_core_big_dextcnt_count(
> + void *obj,
> + int startoff)
> +{
> + struct xfs_dinode *dic;
> +
> + ASSERT(startoff == 0);
> + ASSERT(obj == iocur_top->data);
> + dic = obj;
> +
> + if (dic->di_version == 3 &&
> + (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
(dic->di_flags2 & cpu_to_be64(XFS_DIFLAG2_NREXT64)) ?
(I forget, does userspace's version of those macros elide constants?)
--D
> + return 1;
> + else
> + return 0;
> +}
> +
> +static int
> +inode_core_v3_pad_count(
> + void *obj,
> + int startoff)
> +{
> + struct xfs_dinode *dic;
> +
> + ASSERT(startoff == 0);
> + ASSERT(obj == iocur_top->data);
> + dic = obj;
> +
> + if ((dic->di_version == 3)
> + && !(be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
> + return 8;
> + else
> + return 0;
> +}
> +
> +static int
> +inode_core_v2_pad_count(
> + void *obj,
> + int startoff)
> +{
> + struct xfs_dinode *dic;
> +
> + ASSERT(startoff == 0);
> + ASSERT(obj == iocur_top->data);
> + dic = obj;
> +
> + if (dic->di_version == 3)
> + return 0;
> + else
> + return 6;
> +}
> +
> +static int
> +inode_core_flushiter_count(
> + void *obj,
> + int startoff)
> +{
> + struct xfs_dinode *dic;
> +
> + ASSERT(startoff == 0);
> + ASSERT(obj == iocur_top->data);
> + dic = obj;
> +
> + if (dic->di_version == 3)
> + return 0;
> + else
> + return 1;
> +}
> +
> +static int
> +inode_core_big_aextcnt_count(
> + void *obj,
> + int startoff)
> +{
> + struct xfs_dinode *dic;
> +
> + ASSERT(startoff == 0);
> + ASSERT(obj == iocur_top->data);
> + dic = obj;
> +
> + if (dic->di_version == 3 &&
> + (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
> + return 1;
> + else
> + return 0;
> +}
> +
> +static int
> +inode_core_nrext64_pad_count(
> + void *obj,
> + int startoff)
> +{
> + struct xfs_dinode *dic;
> +
> + ASSERT(startoff == 0);
> + ASSERT(obj == iocur_top->data);
> + dic = obj;
> +
> + if (dic->di_version == 3 &&
> + (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
> + return 1;
> + else
> + return 0;
> +}
> +
> +static int
> +inode_core_nextents_count(
> + void *obj,
> + int startoff)
> +{
> + struct xfs_dinode *dic;
> +
> + ASSERT(startoff == 0);
> + ASSERT(obj == iocur_top->data);
> + dic = obj;
> +
> + if ((dic->di_version == 3)
> + && (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
> + return 0;
> + else
> + return 1;
> +}
> +
> +static int
> +inode_core_anextents_count(
> + void *obj,
> + int startoff)
> +{
> + struct xfs_dinode *dic;
> +
> + ASSERT(startoff == 0);
> + ASSERT(obj == iocur_top->data);
> + dic = obj;
> +
> + if ((dic->di_version == 3)
> + && (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
> + return 0;
> + else
> + return 1;
> +}
> +
> static int
> inode_f(
> int argc,
> diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
> index bdd13ec9..d4c962fc 100644
> --- a/libxfs/xfs_format.h
> +++ b/libxfs/xfs_format.h
> @@ -980,16 +980,30 @@ typedef struct xfs_dinode {
> __be32 di_nlink; /* number of links to file */
> __be16 di_projid_lo; /* lower part of owner's project id */
> __be16 di_projid_hi; /* higher part owner's project id */
> - __u8 di_pad[6]; /* unused, zeroed space */
> - __be16 di_flushiter; /* incremented on flush */
> + union {
> + __be64 di_big_dextcnt; /* NREXT64 data extents */
> + __u8 di_v3_pad[8]; /* !NREXT64 V3 inode zeroed space */
> + struct {
> + __u8 di_v2_pad[6]; /* V2 inode zeroed space */
> + __be16 di_flushiter; /* V2 inode incremented on flush */
> + };
> + };
> xfs_timestamp_t di_atime; /* time last accessed */
> xfs_timestamp_t di_mtime; /* time last modified */
> xfs_timestamp_t di_ctime; /* time created/inode modified */
> __be64 di_size; /* number of bytes in file */
> __be64 di_nblocks; /* # of direct & btree blocks used */
> __be32 di_extsize; /* basic/minimum extent size for file */
> - __be32 di_nextents; /* number of extents in data fork */
> - __be16 di_anextents; /* number of extents in attribute fork*/
> + union {
> + struct {
> + __be32 di_big_aextcnt; /* NREXT64 attr extents */
> + __be16 di_nrext64_pad; /* NREXT64 unused, zero */
> + } __packed;
> + struct {
> + __be32 di_nextents; /* !NREXT64 data extents */
> + __be16 di_anextents; /* !NREXT64 attr extents */
> + } __packed;
> + };
> __u8 di_forkoff; /* attr fork offs, <<3 for 64b align */
> __s8 di_aformat; /* format of attr fork's data */
> __be32 di_dmevmask; /* DMIG event mask */
> diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
> index 9bddf790..5c6de73b 100644
> --- a/libxfs/xfs_inode_buf.c
> +++ b/libxfs/xfs_inode_buf.c
> @@ -276,6 +276,25 @@ xfs_inode_to_disk_ts(
> return ts;
> }
>
> +static inline void
> +xfs_inode_to_disk_iext_counters(
> + struct xfs_inode *ip,
> + struct xfs_dinode *to)
> +{
> + if (xfs_inode_has_nrext64(ip)) {
> + to->di_big_dextcnt = cpu_to_be64(xfs_ifork_nextents(&ip->i_df));
> + to->di_big_aextcnt = cpu_to_be32(xfs_ifork_nextents(ip->i_afp));
> + /*
> + * We might be upgrading the inode to use larger extent counters
> + * than was previously used. Hence zero the unused field.
> + */
> + to->di_nrext64_pad = cpu_to_be16(0);
> + } else {
> + to->di_nextents = cpu_to_be32(xfs_ifork_nextents(&ip->i_df));
> + to->di_anextents = cpu_to_be16(xfs_ifork_nextents(ip->i_afp));
> + }
> +}
> +
> void
> xfs_inode_to_disk(
> struct xfs_inode *ip,
> @@ -293,7 +312,6 @@ xfs_inode_to_disk(
> to->di_projid_lo = cpu_to_be16(ip->i_projid & 0xffff);
> to->di_projid_hi = cpu_to_be16(ip->i_projid >> 16);
>
> - memset(to->di_pad, 0, sizeof(to->di_pad));
> to->di_atime = xfs_inode_to_disk_ts(ip, inode->i_atime);
> to->di_mtime = xfs_inode_to_disk_ts(ip, inode->i_mtime);
> to->di_ctime = xfs_inode_to_disk_ts(ip, inode->i_ctime);
> @@ -304,8 +322,6 @@ xfs_inode_to_disk(
> to->di_size = cpu_to_be64(ip->i_disk_size);
> to->di_nblocks = cpu_to_be64(ip->i_nblocks);
> to->di_extsize = cpu_to_be32(ip->i_extsize);
> - to->di_nextents = cpu_to_be32(xfs_ifork_nextents(&ip->i_df));
> - to->di_anextents = cpu_to_be16(xfs_ifork_nextents(ip->i_afp));
> to->di_forkoff = ip->i_forkoff;
> to->di_aformat = xfs_ifork_format(ip->i_afp);
> to->di_flags = cpu_to_be16(ip->i_diflags);
> @@ -320,11 +336,14 @@ xfs_inode_to_disk(
> to->di_lsn = cpu_to_be64(lsn);
> memset(to->di_pad2, 0, sizeof(to->di_pad2));
> uuid_copy(&to->di_uuid, &ip->i_mount->m_sb.sb_meta_uuid);
> - to->di_flushiter = 0;
> + memset(to->di_v3_pad, 0, sizeof(to->di_v3_pad));
> } else {
> to->di_version = 2;
> to->di_flushiter = cpu_to_be16(ip->i_flushiter);
> + memset(to->di_v2_pad, 0, sizeof(to->di_v2_pad));
> }
> +
> + xfs_inode_to_disk_iext_counters(ip, to);
> }
>
> static xfs_failaddr_t
> diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
> index 7d5f0015..03036a2c 100644
> --- a/libxfs/xfs_inode_fork.h
> +++ b/libxfs/xfs_inode_fork.h
> @@ -156,14 +156,20 @@ static inline xfs_extnum_t
> xfs_dfork_data_extents(
> struct xfs_dinode *dip)
> {
> - return be32_to_cpu(dip->di_nextents);
> + if (xfs_dinode_has_nrext64(dip))
> + return be64_to_cpu(dip->di_big_dextcnt);
> + else
> + return be32_to_cpu(dip->di_nextents);
> }
>
> static inline xfs_extnum_t
> xfs_dfork_attr_extents(
> struct xfs_dinode *dip)
> {
> - return be16_to_cpu(dip->di_anextents);
> + if (xfs_dinode_has_nrext64(dip))
> + return be32_to_cpu(dip->di_big_aextcnt);
> + else
> + return be16_to_cpu(dip->di_anextents);
> }
>
> static inline xfs_extnum_t
> diff --git a/libxfs/xfs_log_format.h b/libxfs/xfs_log_format.h
> index c13608aa..8f8c9869 100644
> --- a/libxfs/xfs_log_format.h
> +++ b/libxfs/xfs_log_format.h
> @@ -388,16 +388,30 @@ struct xfs_log_dinode {
> uint32_t di_nlink; /* number of links to file */
> uint16_t di_projid_lo; /* lower part of owner's project id */
> uint16_t di_projid_hi; /* higher part of owner's project id */
> - uint8_t di_pad[6]; /* unused, zeroed space */
> - uint16_t di_flushiter; /* incremented on flush */
> + union {
> + uint64_t di_big_dextcnt; /* NREXT64 data extents */
> + uint8_t di_v3_pad[8]; /* !NREXT64 V3 inode zeroed space */
> + struct {
> + uint8_t di_v2_pad[6]; /* V2 inode zeroed space */
> + uint16_t di_flushiter; /* V2 inode incremented on flush */
> + };
> + };
> xfs_log_timestamp_t di_atime; /* time last accessed */
> xfs_log_timestamp_t di_mtime; /* time last modified */
> xfs_log_timestamp_t di_ctime; /* time created/inode modified */
> xfs_fsize_t di_size; /* number of bytes in file */
> xfs_rfsblock_t di_nblocks; /* # of direct & btree blocks used */
> xfs_extlen_t di_extsize; /* basic/minimum extent size for file */
> - uint32_t di_nextents; /* number of extents in data fork */
> - uint16_t di_anextents; /* number of extents in attribute fork*/
> + union {
> + struct {
> + uint32_t di_big_aextcnt; /* NREXT64 attr extents */
> + uint16_t di_nrext64_pad; /* NREXT64 unused, zero */
> + } __packed;
> + struct {
> + uint32_t di_nextents; /* !NREXT64 data extents */
> + uint16_t di_anextents; /* !NREXT64 attr extents */
> + } __packed;
> + };
> uint8_t di_forkoff; /* attr fork offs, <<3 for 64b align */
> int8_t di_aformat; /* format of attr fork's data */
> uint32_t di_dmevmask; /* DMIG event mask */
> diff --git a/logprint/log_misc.c b/logprint/log_misc.c
> index 35e926a3..721c2eb5 100644
> --- a/logprint/log_misc.c
> +++ b/logprint/log_misc.c
> @@ -440,6 +440,8 @@ static void
> xlog_print_trans_inode_core(
> struct xfs_log_dinode *ip)
> {
> + xfs_extnum_t nextents;
> +
> printf(_("INODE CORE\n"));
> printf(_("magic 0x%hx mode 0%ho version %d format %d\n"),
> ip->di_magic, ip->di_mode, (int)ip->di_version,
> @@ -450,11 +452,21 @@ xlog_print_trans_inode_core(
> xlog_extract_dinode_ts(ip->di_atime),
> xlog_extract_dinode_ts(ip->di_mtime),
> xlog_extract_dinode_ts(ip->di_ctime));
> - printf(_("size 0x%llx nblocks 0x%llx extsize 0x%x nextents 0x%x\n"),
> +
> + if (ip->di_flags2 & XFS_DIFLAG2_NREXT64)
> + nextents = ip->di_big_dextcnt;
> + else
> + nextents = ip->di_nextents;
> + printf(_("size 0x%llx nblocks 0x%llx extsize 0x%x nextents 0x%lx\n"),
> (unsigned long long)ip->di_size, (unsigned long long)ip->di_nblocks,
> - ip->di_extsize, ip->di_nextents);
> - printf(_("naextents 0x%x forkoff %d dmevmask 0x%x dmstate 0x%hx\n"),
> - ip->di_anextents, (int)ip->di_forkoff, ip->di_dmevmask,
> + ip->di_extsize, nextents);
> +
> + if (ip->di_flags2 & XFS_DIFLAG2_NREXT64)
> + nextents = ip->di_big_aextcnt;
> + else
> + nextents = ip->di_anextents;
> + printf(_("naextents 0x%lx forkoff %d dmevmask 0x%x dmstate 0x%hx\n"),
> + nextents, (int)ip->di_forkoff, ip->di_dmevmask,
> ip->di_dmstate);
> printf(_("flags 0x%x gen 0x%x\n"),
> ip->di_flags, ip->di_gen);
> diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
> index c9c453f6..5e387f38 100644
> --- a/logprint/log_print_all.c
> +++ b/logprint/log_print_all.c
> @@ -240,7 +240,10 @@ STATIC void
> xlog_recover_print_inode_core(
> struct xfs_log_dinode *di)
> {
> - printf(_(" CORE inode:\n"));
> + xfs_extnum_t nextents;
> + xfs_aextnum_t anextents;
> +
> + printf(_(" CORE inode:\n"));
> if (!print_inode)
> return;
> printf(_(" magic:%c%c mode:0x%x ver:%d format:%d\n"),
> @@ -254,10 +257,19 @@ xlog_recover_print_inode_core(
> xlog_extract_dinode_ts(di->di_mtime),
> xlog_extract_dinode_ts(di->di_ctime));
> printf(_(" flushiter:%d\n"), di->di_flushiter);
> +
> + if (di->di_flags2 & XFS_DIFLAG2_NREXT64) {
> + nextents = di->di_big_dextcnt;
> + anextents = di->di_big_aextcnt;
> + } else {
> + nextents = di->di_nextents;
> + anextents = di->di_anextents;
> + }
> +
> printf(_(" size:0x%llx nblks:0x%llx exsize:%d "
> - "nextents:%d anextents:%d\n"), (unsigned long long)
> + "nextents:%lu anextents:%u\n"), (unsigned long long)
> di->di_size, (unsigned long long)di->di_nblocks,
> - di->di_extsize, di->di_nextents, (int)di->di_anextents);
> + di->di_extsize, nextents, anextents);
> printf(_(" forkoff:%d dmevmask:0x%x dmstate:%d flags:0x%x "
> "gen:%u\n"),
> (int)di->di_forkoff, di->di_dmevmask, (int)di->di_dmstate,
> diff --git a/repair/dinode.c b/repair/dinode.c
> index 0df84e48..3be2e1d5 100644
> --- a/repair/dinode.c
> +++ b/repair/dinode.c
> @@ -71,7 +71,12 @@ _("would have cleared inode %" PRIu64 " attributes\n"), ino_num);
> if (xfs_dfork_attr_extents(dino) != 0) {
> if (no_modify)
> return(1);
> - dino->di_anextents = cpu_to_be16(0);
> +
> + if (xfs_dinode_has_nrext64(dino))
> + dino->di_big_aextcnt = 0;
> + else
> + dino->di_anextents = 0;
> +
> }
>
> if (dino->di_aformat != XFS_DINODE_FMT_EXTENTS) {
> @@ -1818,7 +1823,10 @@ _("too many data fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
> do_warn(
> _("correcting nextents for inode %" PRIu64 ", was %lu - counted %" PRIu64 "\n"),
> lino, dnextents, nextents);
> - dino->di_nextents = cpu_to_be32(nextents);
> + if (xfs_dinode_has_nrext64(dino))
> + dino->di_big_dextcnt = cpu_to_be64(nextents);
> + else
> + dino->di_nextents = cpu_to_be32(nextents);
> *dirty = 1;
> } else {
> do_warn(
> @@ -1841,7 +1849,11 @@ _("too many attr fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
> do_warn(
> _("correcting anextents for inode %" PRIu64 ", was %lu - counted %" PRIu64 "\n"),
> lino, dnextents, anextents);
> - dino->di_anextents = cpu_to_be16(anextents);
> + if (xfs_dinode_has_nrext64(dino))
> + dino->di_big_aextcnt = cpu_to_be32(anextents);
> + else
> + dino->di_anextents = cpu_to_be16(anextents);
> +
> *dirty = 1;
> } else {
> do_warn(
> --
> 2.30.2
>
^ permalink raw reply [flat|nested] 34+ messages in thread* Re: [PATCH V4 13/20] xfsprogs: Introduce per-inode 64-bit extent counters
2022-01-05 1:12 ` Darrick J. Wong
@ 2022-01-07 16:10 ` Chandan Babu R
0 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2022-01-07 16:10 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: linux-xfs, david, Dave Chinner
On 05 Jan 2022 at 06:42, Darrick J. Wong wrote:
> On Tue, Dec 14, 2021 at 02:18:04PM +0530, Chandan Babu R wrote:
>> This commit introduces new fields in the on-disk inode format to support
>> 64-bit data fork extent counters and 32-bit attribute fork extent
>> counters. The new fields will be used only when an inode has
>> XFS_DIFLAG2_NREXT64 flag set. Otherwise we continue to use the regular 32-bit
>> data fork extent counters and 16-bit attribute fork extent counters.
>>
>> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
>> Suggested-by: Dave Chinner <dchinner@redhat.com>
>> ---
>> db/field.c | 4 -
>> db/field.h | 2 -
>> db/inode.c | 170 ++++++++++++++++++++++++++++++++++++++-
>> libxfs/xfs_format.h | 22 ++++-
>> libxfs/xfs_inode_buf.c | 27 ++++++-
>> libxfs/xfs_inode_fork.h | 10 ++-
>> libxfs/xfs_log_format.h | 22 ++++-
>> logprint/log_misc.c | 20 ++++-
>> logprint/log_print_all.c | 18 ++++-
>> repair/dinode.c | 18 ++++-
>> 10 files changed, 279 insertions(+), 34 deletions(-)
>>
>> diff --git a/db/field.c b/db/field.c
>> index 51268938..1e274ffc 100644
>> --- a/db/field.c
>> +++ b/db/field.c
>> @@ -25,8 +25,6 @@
>> #include "symlink.h"
>>
>> const ftattr_t ftattrtab[] = {
>> - { FLDT_AEXTNUM, "aextnum", fp_num, "%d", SI(bitsz(xfs_aextnum_t)),
>> - FTARG_SIGNED, NULL, NULL },
>> { FLDT_AGBLOCK, "agblock", fp_num, "%u", SI(bitsz(xfs_agblock_t)),
>> FTARG_DONULL, fa_agblock, NULL },
>> { FLDT_AGBLOCKNZ, "agblocknz", fp_num, "%u", SI(bitsz(xfs_agblock_t)),
>> @@ -300,8 +298,6 @@ const ftattr_t ftattrtab[] = {
>> FTARG_DONULL, fa_drtbno, NULL },
>> { FLDT_EXTLEN, "extlen", fp_num, "%u", SI(bitsz(xfs_extlen_t)), 0, NULL,
>> NULL },
>> - { FLDT_EXTNUM, "extnum", fp_num, "%d", SI(bitsz(xfs_extnum_t)),
>> - FTARG_SIGNED, NULL, NULL },
>> { FLDT_FSIZE, "fsize", fp_num, "%lld", SI(bitsz(xfs_fsize_t)),
>> FTARG_SIGNED, NULL, NULL },
>> { FLDT_INO, "ino", fp_num, "%llu", SI(bitsz(xfs_ino_t)), FTARG_DONULL,
>> diff --git a/db/field.h b/db/field.h
>> index 387c189e..614fd0ab 100644
>> --- a/db/field.h
>> +++ b/db/field.h
>> @@ -5,7 +5,6 @@
>> */
>>
>> typedef enum fldt {
>> - FLDT_AEXTNUM,
>> FLDT_AGBLOCK,
>> FLDT_AGBLOCKNZ,
>> FLDT_AGF,
>> @@ -143,7 +142,6 @@ typedef enum fldt {
>> FLDT_DRFSBNO,
>> FLDT_DRTBNO,
>> FLDT_EXTLEN,
>> - FLDT_EXTNUM,
>> FLDT_FSIZE,
>> FLDT_INO,
>> FLDT_INOBT,
>> diff --git a/db/inode.c b/db/inode.c
>> index b1f92d36..226797be 100644
>> --- a/db/inode.c
>> +++ b/db/inode.c
>> @@ -27,6 +27,14 @@ static int inode_core_nlinkv2_count(void *obj, int startoff);
>> static int inode_core_onlink_count(void *obj, int startoff);
>> static int inode_core_projid_count(void *obj, int startoff);
>> static int inode_core_nlinkv1_count(void *obj, int startoff);
>> +static int inode_core_big_dextcnt_count(void *obj, int startoff);
>> +static int inode_core_v3_pad_count(void *obj, int startoff);
>> +static int inode_core_v2_pad_count(void *obj, int startoff);
>> +static int inode_core_flushiter_count(void *obj, int startoff);
>> +static int inode_core_big_aextcnt_count(void *obj, int startoff);
>> +static int inode_core_nrext64_pad_count(void *obj, int startoff);
>> +static int inode_core_nextents_count(void *obj, int startoff);
>> +static int inode_core_anextents_count(void *obj, int startoff);
>> static int inode_f(int argc, char **argv);
>> static int inode_u_offset(void *obj, int startoff, int idx);
>> static int inode_u_bmbt_count(void *obj, int startoff);
>> @@ -90,18 +98,30 @@ const field_t inode_core_flds[] = {
>> inode_core_projid_count, FLD_COUNT, TYP_NONE },
>> { "projid_hi", FLDT_UINT16D, OI(COFF(projid_hi)),
>> inode_core_projid_count, FLD_COUNT, TYP_NONE },
>> - { "pad", FLDT_UINT8X, OI(OFF(pad)), CI(6), FLD_ARRAY|FLD_SKIPALL, TYP_NONE },
>> + { "big_dextcnt", FLDT_UINT64D, OI(COFF(big_dextcnt)),
>
> Any reason we don't leave the display names as nextents and naextents?
Using existing names to display extent counts would require an xfs_db user to
additionally check the status of XFS_DIFLAG2_NREXT64 flag in order to
determine the exact source of the extent count (large extent counter vs small
extent counter).
This was the reason to print the large extent counters with different field
names.
Come to think of it, Most of the times an xfs_db user would generally be
interested in the extent counter value rather than the exact field from which
the count has been obtained from. Hence in the interest of optimizing for the
general case, I will change the code to continue to use nextents/naextents
names.
> I think fstests has a fair amount of hardcoded xfs_db output, right?
>
Actually xfs/298 is the only test which refers to nextents field directly,
_scratch_xfs_db -c "inode $inode" -c "p core.nextents"
Other tests use the helper function _scratch_get_iext_count() to obtain the
value of extent counters.
I had extended _scratch_get_iext_count() to obtain the correct data/attr fork
extent counters based on whether the per-inode flag XFS_DIFLAG2_NREXT64 is set
or not. This won't be required when xfs_db uses the same names to print the
extent counters.
>> + inode_core_big_dextcnt_count, FLD_COUNT, TYP_NONE },
>> + { "v3_pad", FLDT_UINT8X, OI(OFF(v3_pad)),
>> + inode_core_v3_pad_count, FLD_ARRAY|FLD_COUNT|FLD_SKIPALL, TYP_NONE },
>> + { "v2_pad", FLDT_UINT8X, OI(OFF(v2_pad)),
>> + inode_core_v2_pad_count, FLD_ARRAY|FLD_COUNT|FLD_SKIPALL, TYP_NONE },
>> { "uid", FLDT_UINT32D, OI(COFF(uid)), C1, 0, TYP_NONE },
>> { "gid", FLDT_UINT32D, OI(COFF(gid)), C1, 0, TYP_NONE },
>> - { "flushiter", FLDT_UINT16D, OI(COFF(flushiter)), C1, 0, TYP_NONE },
>> + { "flushiter", FLDT_UINT16D, OI(COFF(flushiter)),
>> + inode_core_flushiter_count, FLD_COUNT, TYP_NONE },
>> { "atime", FLDT_TIMESTAMP, OI(COFF(atime)), C1, 0, TYP_NONE },
>> { "mtime", FLDT_TIMESTAMP, OI(COFF(mtime)), C1, 0, TYP_NONE },
>> { "ctime", FLDT_TIMESTAMP, OI(COFF(ctime)), C1, 0, TYP_NONE },
>> { "size", FLDT_FSIZE, OI(COFF(size)), C1, 0, TYP_NONE },
>> { "nblocks", FLDT_DRFSBNO, OI(COFF(nblocks)), C1, 0, TYP_NONE },
>> { "extsize", FLDT_EXTLEN, OI(COFF(extsize)), C1, 0, TYP_NONE },
>> - { "nextents", FLDT_EXTNUM, OI(COFF(nextents)), C1, 0, TYP_NONE },
>> - { "naextents", FLDT_AEXTNUM, OI(COFF(anextents)), C1, 0, TYP_NONE },
>> + { "big_aextcnt", FLDT_UINT32D, OI(COFF(big_aextcnt)),
>> + inode_core_big_aextcnt_count, FLD_COUNT, TYP_NONE },
>> + { "nrext64_pad", FLDT_UINT16D, OI(COFF(nrext64_pad)),
>> + inode_core_nrext64_pad_count, FLD_COUNT, TYP_NONE },
>> + { "nextents", FLDT_UINT32D, OI(COFF(nextents)),
>> + inode_core_nextents_count, FLD_COUNT, TYP_NONE },
>> + { "naextents", FLDT_UINT16D, OI(COFF(anextents)),
>> + inode_core_anextents_count, FLD_COUNT, TYP_NONE },
>> { "forkoff", FLDT_UINT8D, OI(COFF(forkoff)), C1, 0, TYP_NONE },
>> { "aformat", FLDT_DINODE_FMT, OI(COFF(aformat)), C1, 0, TYP_NONE },
>> { "dmevmask", FLDT_UINT32X, OI(COFF(dmevmask)), C1, 0, TYP_NONE },
>> @@ -403,6 +423,148 @@ inode_core_projid_count(
>> return dic->di_version >= 2;
>> }
>>
>> +static int
>> +inode_core_big_dextcnt_count(
>> + void *obj,
>> + int startoff)
>> +{
>> + struct xfs_dinode *dic;
>> +
>> + ASSERT(startoff == 0);
>> + ASSERT(obj == iocur_top->data);
>> + dic = obj;
>> +
>> + if (dic->di_version == 3 &&
>> + (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
>
> (dic->di_flags2 & cpu_to_be64(XFS_DIFLAG2_NREXT64)) ?
>
> (I forget, does userspace's version of those macros elide constants?)
>
> --D
>
>> + return 1;
>> + else
>> + return 0;
>> +}
>> +
>> +static int
>> +inode_core_v3_pad_count(
>> + void *obj,
>> + int startoff)
>> +{
>> + struct xfs_dinode *dic;
>> +
>> + ASSERT(startoff == 0);
>> + ASSERT(obj == iocur_top->data);
>> + dic = obj;
>> +
>> + if ((dic->di_version == 3)
>> + && !(be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
>> + return 8;
>> + else
>> + return 0;
>> +}
>> +
>> +static int
>> +inode_core_v2_pad_count(
>> + void *obj,
>> + int startoff)
>> +{
>> + struct xfs_dinode *dic;
>> +
>> + ASSERT(startoff == 0);
>> + ASSERT(obj == iocur_top->data);
>> + dic = obj;
>> +
>> + if (dic->di_version == 3)
>> + return 0;
>> + else
>> + return 6;
>> +}
>> +
>> +static int
>> +inode_core_flushiter_count(
>> + void *obj,
>> + int startoff)
>> +{
>> + struct xfs_dinode *dic;
>> +
>> + ASSERT(startoff == 0);
>> + ASSERT(obj == iocur_top->data);
>> + dic = obj;
>> +
>> + if (dic->di_version == 3)
>> + return 0;
>> + else
>> + return 1;
>> +}
>> +
>> +static int
>> +inode_core_big_aextcnt_count(
>> + void *obj,
>> + int startoff)
>> +{
>> + struct xfs_dinode *dic;
>> +
>> + ASSERT(startoff == 0);
>> + ASSERT(obj == iocur_top->data);
>> + dic = obj;
>> +
>> + if (dic->di_version == 3 &&
>> + (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
>> + return 1;
>> + else
>> + return 0;
>> +}
>> +
>> +static int
>> +inode_core_nrext64_pad_count(
>> + void *obj,
>> + int startoff)
>> +{
>> + struct xfs_dinode *dic;
>> +
>> + ASSERT(startoff == 0);
>> + ASSERT(obj == iocur_top->data);
>> + dic = obj;
>> +
>> + if (dic->di_version == 3 &&
>> + (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
>> + return 1;
>> + else
>> + return 0;
>> +}
>> +
>> +static int
>> +inode_core_nextents_count(
>> + void *obj,
>> + int startoff)
>> +{
>> + struct xfs_dinode *dic;
>> +
>> + ASSERT(startoff == 0);
>> + ASSERT(obj == iocur_top->data);
>> + dic = obj;
>> +
>> + if ((dic->di_version == 3)
>> + && (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
>> + return 0;
>> + else
>> + return 1;
>> +}
>> +
>> +static int
>> +inode_core_anextents_count(
>> + void *obj,
>> + int startoff)
>> +{
>> + struct xfs_dinode *dic;
>> +
>> + ASSERT(startoff == 0);
>> + ASSERT(obj == iocur_top->data);
>> + dic = obj;
>> +
>> + if ((dic->di_version == 3)
>> + && (be64_to_cpu(dic->di_flags2) & XFS_DIFLAG2_NREXT64))
>> + return 0;
>> + else
>> + return 1;
>> +}
>> +
>> static int
>> inode_f(
>> int argc,
>> diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
>> index bdd13ec9..d4c962fc 100644
>> --- a/libxfs/xfs_format.h
>> +++ b/libxfs/xfs_format.h
>> @@ -980,16 +980,30 @@ typedef struct xfs_dinode {
>> __be32 di_nlink; /* number of links to file */
>> __be16 di_projid_lo; /* lower part of owner's project id */
>> __be16 di_projid_hi; /* higher part owner's project id */
>> - __u8 di_pad[6]; /* unused, zeroed space */
>> - __be16 di_flushiter; /* incremented on flush */
>> + union {
>> + __be64 di_big_dextcnt; /* NREXT64 data extents */
>> + __u8 di_v3_pad[8]; /* !NREXT64 V3 inode zeroed space */
>> + struct {
>> + __u8 di_v2_pad[6]; /* V2 inode zeroed space */
>> + __be16 di_flushiter; /* V2 inode incremented on flush */
>> + };
>> + };
>> xfs_timestamp_t di_atime; /* time last accessed */
>> xfs_timestamp_t di_mtime; /* time last modified */
>> xfs_timestamp_t di_ctime; /* time created/inode modified */
>> __be64 di_size; /* number of bytes in file */
>> __be64 di_nblocks; /* # of direct & btree blocks used */
>> __be32 di_extsize; /* basic/minimum extent size for file */
>> - __be32 di_nextents; /* number of extents in data fork */
>> - __be16 di_anextents; /* number of extents in attribute fork*/
>> + union {
>> + struct {
>> + __be32 di_big_aextcnt; /* NREXT64 attr extents */
>> + __be16 di_nrext64_pad; /* NREXT64 unused, zero */
>> + } __packed;
>> + struct {
>> + __be32 di_nextents; /* !NREXT64 data extents */
>> + __be16 di_anextents; /* !NREXT64 attr extents */
>> + } __packed;
>> + };
>> __u8 di_forkoff; /* attr fork offs, <<3 for 64b align */
>> __s8 di_aformat; /* format of attr fork's data */
>> __be32 di_dmevmask; /* DMIG event mask */
>> diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
>> index 9bddf790..5c6de73b 100644
>> --- a/libxfs/xfs_inode_buf.c
>> +++ b/libxfs/xfs_inode_buf.c
>> @@ -276,6 +276,25 @@ xfs_inode_to_disk_ts(
>> return ts;
>> }
>>
>> +static inline void
>> +xfs_inode_to_disk_iext_counters(
>> + struct xfs_inode *ip,
>> + struct xfs_dinode *to)
>> +{
>> + if (xfs_inode_has_nrext64(ip)) {
>> + to->di_big_dextcnt = cpu_to_be64(xfs_ifork_nextents(&ip->i_df));
>> + to->di_big_aextcnt = cpu_to_be32(xfs_ifork_nextents(ip->i_afp));
>> + /*
>> + * We might be upgrading the inode to use larger extent counters
>> + * than was previously used. Hence zero the unused field.
>> + */
>> + to->di_nrext64_pad = cpu_to_be16(0);
>> + } else {
>> + to->di_nextents = cpu_to_be32(xfs_ifork_nextents(&ip->i_df));
>> + to->di_anextents = cpu_to_be16(xfs_ifork_nextents(ip->i_afp));
>> + }
>> +}
>> +
>> void
>> xfs_inode_to_disk(
>> struct xfs_inode *ip,
>> @@ -293,7 +312,6 @@ xfs_inode_to_disk(
>> to->di_projid_lo = cpu_to_be16(ip->i_projid & 0xffff);
>> to->di_projid_hi = cpu_to_be16(ip->i_projid >> 16);
>>
>> - memset(to->di_pad, 0, sizeof(to->di_pad));
>> to->di_atime = xfs_inode_to_disk_ts(ip, inode->i_atime);
>> to->di_mtime = xfs_inode_to_disk_ts(ip, inode->i_mtime);
>> to->di_ctime = xfs_inode_to_disk_ts(ip, inode->i_ctime);
>> @@ -304,8 +322,6 @@ xfs_inode_to_disk(
>> to->di_size = cpu_to_be64(ip->i_disk_size);
>> to->di_nblocks = cpu_to_be64(ip->i_nblocks);
>> to->di_extsize = cpu_to_be32(ip->i_extsize);
>> - to->di_nextents = cpu_to_be32(xfs_ifork_nextents(&ip->i_df));
>> - to->di_anextents = cpu_to_be16(xfs_ifork_nextents(ip->i_afp));
>> to->di_forkoff = ip->i_forkoff;
>> to->di_aformat = xfs_ifork_format(ip->i_afp);
>> to->di_flags = cpu_to_be16(ip->i_diflags);
>> @@ -320,11 +336,14 @@ xfs_inode_to_disk(
>> to->di_lsn = cpu_to_be64(lsn);
>> memset(to->di_pad2, 0, sizeof(to->di_pad2));
>> uuid_copy(&to->di_uuid, &ip->i_mount->m_sb.sb_meta_uuid);
>> - to->di_flushiter = 0;
>> + memset(to->di_v3_pad, 0, sizeof(to->di_v3_pad));
>> } else {
>> to->di_version = 2;
>> to->di_flushiter = cpu_to_be16(ip->i_flushiter);
>> + memset(to->di_v2_pad, 0, sizeof(to->di_v2_pad));
>> }
>> +
>> + xfs_inode_to_disk_iext_counters(ip, to);
>> }
>>
>> static xfs_failaddr_t
>> diff --git a/libxfs/xfs_inode_fork.h b/libxfs/xfs_inode_fork.h
>> index 7d5f0015..03036a2c 100644
>> --- a/libxfs/xfs_inode_fork.h
>> +++ b/libxfs/xfs_inode_fork.h
>> @@ -156,14 +156,20 @@ static inline xfs_extnum_t
>> xfs_dfork_data_extents(
>> struct xfs_dinode *dip)
>> {
>> - return be32_to_cpu(dip->di_nextents);
>> + if (xfs_dinode_has_nrext64(dip))
>> + return be64_to_cpu(dip->di_big_dextcnt);
>> + else
>> + return be32_to_cpu(dip->di_nextents);
>> }
>>
>> static inline xfs_extnum_t
>> xfs_dfork_attr_extents(
>> struct xfs_dinode *dip)
>> {
>> - return be16_to_cpu(dip->di_anextents);
>> + if (xfs_dinode_has_nrext64(dip))
>> + return be32_to_cpu(dip->di_big_aextcnt);
>> + else
>> + return be16_to_cpu(dip->di_anextents);
>> }
>>
>> static inline xfs_extnum_t
>> diff --git a/libxfs/xfs_log_format.h b/libxfs/xfs_log_format.h
>> index c13608aa..8f8c9869 100644
>> --- a/libxfs/xfs_log_format.h
>> +++ b/libxfs/xfs_log_format.h
>> @@ -388,16 +388,30 @@ struct xfs_log_dinode {
>> uint32_t di_nlink; /* number of links to file */
>> uint16_t di_projid_lo; /* lower part of owner's project id */
>> uint16_t di_projid_hi; /* higher part of owner's project id */
>> - uint8_t di_pad[6]; /* unused, zeroed space */
>> - uint16_t di_flushiter; /* incremented on flush */
>> + union {
>> + uint64_t di_big_dextcnt; /* NREXT64 data extents */
>> + uint8_t di_v3_pad[8]; /* !NREXT64 V3 inode zeroed space */
>> + struct {
>> + uint8_t di_v2_pad[6]; /* V2 inode zeroed space */
>> + uint16_t di_flushiter; /* V2 inode incremented on flush */
>> + };
>> + };
>> xfs_log_timestamp_t di_atime; /* time last accessed */
>> xfs_log_timestamp_t di_mtime; /* time last modified */
>> xfs_log_timestamp_t di_ctime; /* time created/inode modified */
>> xfs_fsize_t di_size; /* number of bytes in file */
>> xfs_rfsblock_t di_nblocks; /* # of direct & btree blocks used */
>> xfs_extlen_t di_extsize; /* basic/minimum extent size for file */
>> - uint32_t di_nextents; /* number of extents in data fork */
>> - uint16_t di_anextents; /* number of extents in attribute fork*/
>> + union {
>> + struct {
>> + uint32_t di_big_aextcnt; /* NREXT64 attr extents */
>> + uint16_t di_nrext64_pad; /* NREXT64 unused, zero */
>> + } __packed;
>> + struct {
>> + uint32_t di_nextents; /* !NREXT64 data extents */
>> + uint16_t di_anextents; /* !NREXT64 attr extents */
>> + } __packed;
>> + };
>> uint8_t di_forkoff; /* attr fork offs, <<3 for 64b align */
>> int8_t di_aformat; /* format of attr fork's data */
>> uint32_t di_dmevmask; /* DMIG event mask */
>> diff --git a/logprint/log_misc.c b/logprint/log_misc.c
>> index 35e926a3..721c2eb5 100644
>> --- a/logprint/log_misc.c
>> +++ b/logprint/log_misc.c
>> @@ -440,6 +440,8 @@ static void
>> xlog_print_trans_inode_core(
>> struct xfs_log_dinode *ip)
>> {
>> + xfs_extnum_t nextents;
>> +
>> printf(_("INODE CORE\n"));
>> printf(_("magic 0x%hx mode 0%ho version %d format %d\n"),
>> ip->di_magic, ip->di_mode, (int)ip->di_version,
>> @@ -450,11 +452,21 @@ xlog_print_trans_inode_core(
>> xlog_extract_dinode_ts(ip->di_atime),
>> xlog_extract_dinode_ts(ip->di_mtime),
>> xlog_extract_dinode_ts(ip->di_ctime));
>> - printf(_("size 0x%llx nblocks 0x%llx extsize 0x%x nextents 0x%x\n"),
>> +
>> + if (ip->di_flags2 & XFS_DIFLAG2_NREXT64)
>> + nextents = ip->di_big_dextcnt;
>> + else
>> + nextents = ip->di_nextents;
>> + printf(_("size 0x%llx nblocks 0x%llx extsize 0x%x nextents 0x%lx\n"),
>> (unsigned long long)ip->di_size, (unsigned long long)ip->di_nblocks,
>> - ip->di_extsize, ip->di_nextents);
>> - printf(_("naextents 0x%x forkoff %d dmevmask 0x%x dmstate 0x%hx\n"),
>> - ip->di_anextents, (int)ip->di_forkoff, ip->di_dmevmask,
>> + ip->di_extsize, nextents);
>> +
>> + if (ip->di_flags2 & XFS_DIFLAG2_NREXT64)
>> + nextents = ip->di_big_aextcnt;
>> + else
>> + nextents = ip->di_anextents;
>> + printf(_("naextents 0x%lx forkoff %d dmevmask 0x%x dmstate 0x%hx\n"),
>> + nextents, (int)ip->di_forkoff, ip->di_dmevmask,
>> ip->di_dmstate);
>> printf(_("flags 0x%x gen 0x%x\n"),
>> ip->di_flags, ip->di_gen);
>> diff --git a/logprint/log_print_all.c b/logprint/log_print_all.c
>> index c9c453f6..5e387f38 100644
>> --- a/logprint/log_print_all.c
>> +++ b/logprint/log_print_all.c
>> @@ -240,7 +240,10 @@ STATIC void
>> xlog_recover_print_inode_core(
>> struct xfs_log_dinode *di)
>> {
>> - printf(_(" CORE inode:\n"));
>> + xfs_extnum_t nextents;
>> + xfs_aextnum_t anextents;
>> +
>> + printf(_(" CORE inode:\n"));
>> if (!print_inode)
>> return;
>> printf(_(" magic:%c%c mode:0x%x ver:%d format:%d\n"),
>> @@ -254,10 +257,19 @@ xlog_recover_print_inode_core(
>> xlog_extract_dinode_ts(di->di_mtime),
>> xlog_extract_dinode_ts(di->di_ctime));
>> printf(_(" flushiter:%d\n"), di->di_flushiter);
>> +
>> + if (di->di_flags2 & XFS_DIFLAG2_NREXT64) {
>> + nextents = di->di_big_dextcnt;
>> + anextents = di->di_big_aextcnt;
>> + } else {
>> + nextents = di->di_nextents;
>> + anextents = di->di_anextents;
>> + }
>> +
>> printf(_(" size:0x%llx nblks:0x%llx exsize:%d "
>> - "nextents:%d anextents:%d\n"), (unsigned long long)
>> + "nextents:%lu anextents:%u\n"), (unsigned long long)
>> di->di_size, (unsigned long long)di->di_nblocks,
>> - di->di_extsize, di->di_nextents, (int)di->di_anextents);
>> + di->di_extsize, nextents, anextents);
>> printf(_(" forkoff:%d dmevmask:0x%x dmstate:%d flags:0x%x "
>> "gen:%u\n"),
>> (int)di->di_forkoff, di->di_dmevmask, (int)di->di_dmstate,
>> diff --git a/repair/dinode.c b/repair/dinode.c
>> index 0df84e48..3be2e1d5 100644
>> --- a/repair/dinode.c
>> +++ b/repair/dinode.c
>> @@ -71,7 +71,12 @@ _("would have cleared inode %" PRIu64 " attributes\n"), ino_num);
>> if (xfs_dfork_attr_extents(dino) != 0) {
>> if (no_modify)
>> return(1);
>> - dino->di_anextents = cpu_to_be16(0);
>> +
>> + if (xfs_dinode_has_nrext64(dino))
>> + dino->di_big_aextcnt = 0;
>> + else
>> + dino->di_anextents = 0;
>> +
>> }
>>
>> if (dino->di_aformat != XFS_DINODE_FMT_EXTENTS) {
>> @@ -1818,7 +1823,10 @@ _("too many data fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
>> do_warn(
>> _("correcting nextents for inode %" PRIu64 ", was %lu - counted %" PRIu64 "\n"),
>> lino, dnextents, nextents);
>> - dino->di_nextents = cpu_to_be32(nextents);
>> + if (xfs_dinode_has_nrext64(dino))
>> + dino->di_big_dextcnt = cpu_to_be64(nextents);
>> + else
>> + dino->di_nextents = cpu_to_be32(nextents);
>> *dirty = 1;
>> } else {
>> do_warn(
>> @@ -1841,7 +1849,11 @@ _("too many attr fork extents (%" PRIu64 ") in inode %" PRIu64 "\n"),
>> do_warn(
>> _("correcting anextents for inode %" PRIu64 ", was %lu - counted %" PRIu64 "\n"),
>> lino, dnextents, anextents);
>> - dino->di_anextents = cpu_to_be16(anextents);
>> + if (xfs_dinode_has_nrext64(dino))
>> + dino->di_big_aextcnt = cpu_to_be32(anextents);
>> + else
>> + dino->di_anextents = cpu_to_be16(anextents);
>> +
>> *dirty = 1;
>> } else {
>> do_warn(
>> --
>> 2.30.2
>>
--
chandan
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH V4 14/20] xfsprogs: Conditionally upgrade existing inodes to use 64-bit extent counters
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (12 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 13/20] xfsprogs: Introduce per-inode 64-bit extent counters Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2021-12-14 8:48 ` [PATCH V4 15/20] xfsprogs: Enable bulkstat ioctl to support " Chandan Babu R
` (5 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
This commit upgrades inodes to use 64-bit extent counters when they are read
from disk. Inodes are upgraded only when the filesystem instance has
XFS_SB_FEAT_INCOMPAT_NREXT64 incompat flag set.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_inode_buf.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
index 5c6de73b..a484e14b 100644
--- a/libxfs/xfs_inode_buf.c
+++ b/libxfs/xfs_inode_buf.c
@@ -250,6 +250,12 @@ xfs_inode_from_disk(
}
if (xfs_is_reflink_inode(ip))
xfs_ifork_init_cow(ip);
+
+ if ((from->di_version == 3) &&
+ xfs_sb_version_hasnrext64(&ip->i_mount->m_sb) &&
+ !xfs_dinode_has_nrext64(from))
+ ip->i_diflags2 |= XFS_DIFLAG2_NREXT64;
+
return 0;
out_destroy_data_fork:
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 15/20] xfsprogs: Enable bulkstat ioctl to support 64-bit extent counters
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (13 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 14/20] xfsprogs: Conditionally upgrade existing inodes to use " Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2022-01-05 1:13 ` Darrick J. Wong
2021-12-14 8:48 ` [PATCH V4 16/20] xfsprogs: Add XFS_SB_FEAT_INCOMPAT_NREXT64 to the list of supported flags Chandan Babu R
` (4 subsequent siblings)
19 siblings, 1 reply; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
This commit adds support to libfrog to enable reporting 64-bit extent counters
to its users. In order to do so, bulkstat ioctl is now invoked with the newly
introduced XFS_BULK_IREQ_NREXT64 flag if the underlying filesystem's geometry
supports 64-bit extent counters.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
fsr/xfs_fsr.c | 4 ++--
io/bulkstat.c | 1 +
libfrog/bulkstat.c | 29 +++++++++++++++++++++++++++--
libxfs/xfs_fs.h | 12 ++++++++----
4 files changed, 38 insertions(+), 8 deletions(-)
diff --git a/fsr/xfs_fsr.c b/fsr/xfs_fsr.c
index 6cf8bfb7..ba02506d 100644
--- a/fsr/xfs_fsr.c
+++ b/fsr/xfs_fsr.c
@@ -590,7 +590,7 @@ cmp(const void *s1, const void *s2)
(bs1->bs_version == XFS_BULKSTAT_VERSION_V5 &&
bs2->bs_version == XFS_BULKSTAT_VERSION_V5));
- return (bs2->bs_extents - bs1->bs_extents);
+ return (bs2->bs_extents64 - bs1->bs_extents64);
}
/*
@@ -655,7 +655,7 @@ fsrfs(char *mntdir, xfs_ino_t startino, int targetrange)
for (p = buf, endp = (buf + buflenout); p < endp ; p++) {
/* Do some obvious checks now */
if (((p->bs_mode & S_IFMT) != S_IFREG) ||
- (p->bs_extents < 2))
+ (p->bs_extents64 < 2))
continue;
ret = -xfrog_bulkstat_v5_to_v1(&fsxfd, &bs1, p);
diff --git a/io/bulkstat.c b/io/bulkstat.c
index 201470b2..0c9a2b02 100644
--- a/io/bulkstat.c
+++ b/io/bulkstat.c
@@ -57,6 +57,7 @@ dump_bulkstat(
printf("\tbs_sick = 0x%"PRIx16"\n", bstat->bs_sick);
printf("\tbs_checked = 0x%"PRIx16"\n", bstat->bs_checked);
printf("\tbs_mode = 0%"PRIo16"\n", bstat->bs_mode);
+ printf("\tbs_extents64 = %"PRIu64"\n", bstat->bs_extents64);
};
static void
diff --git a/libfrog/bulkstat.c b/libfrog/bulkstat.c
index 195f6ea0..0a90947f 100644
--- a/libfrog/bulkstat.c
+++ b/libfrog/bulkstat.c
@@ -56,6 +56,9 @@ xfrog_bulkstat_single5(
if (flags & ~(XFS_BULK_IREQ_SPECIAL))
return -EINVAL;
+ if (xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64)
+ flags |= XFS_BULK_IREQ_NREXT64;
+
ret = xfrog_bulkstat_alloc_req(1, ino, &req);
if (ret)
return ret;
@@ -73,6 +76,12 @@ xfrog_bulkstat_single5(
}
memcpy(bulkstat, req->bulkstat, sizeof(struct xfs_bulkstat));
+
+ if (!(xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64)) {
+ bulkstat->bs_extents64 = bulkstat->bs_extents;
+ bulkstat->bs_extents = 0;
+ }
+
free:
free(req);
return ret;
@@ -129,6 +138,7 @@ xfrog_bulkstat_single(
switch (error) {
case -EOPNOTSUPP:
case -ENOTTY:
+ assert(!(xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64));
xfd->flags |= XFROG_FLAG_BULKSTAT_FORCE_V1;
break;
}
@@ -259,10 +269,23 @@ xfrog_bulkstat5(
struct xfs_bulkstat_req *req)
{
int ret;
+ int i;
+
+ if (xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64)
+ req->hdr.flags |= XFS_BULK_IREQ_NREXT64;
ret = ioctl(xfd->fd, XFS_IOC_BULKSTAT, req);
if (ret)
return -errno;
+
+ if (!(xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64)) {
+ for (i = 0; i < req->hdr.ocount; i++) {
+ req->bulkstat[i].bs_extents64 =
+ req->bulkstat[i].bs_extents;
+ req->bulkstat[i].bs_extents = 0;
+ }
+ }
+
return 0;
}
@@ -316,6 +339,7 @@ xfrog_bulkstat(
switch (error) {
case -EOPNOTSUPP:
case -ENOTTY:
+ assert(!(xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64));
xfd->flags |= XFROG_FLAG_BULKSTAT_FORCE_V1;
break;
}
@@ -342,6 +366,7 @@ xfrog_bulkstat_v5_to_v1(
const struct xfs_bulkstat *bs5)
{
if (bs5->bs_aextents > UINT16_MAX ||
+ bs5->bs_extents64 > INT32_MAX ||
cvt_off_fsb_to_b(xfd, bs5->bs_extsize_blks) > UINT32_MAX ||
cvt_off_fsb_to_b(xfd, bs5->bs_cowextsize_blks) > UINT32_MAX ||
time_too_big(bs5->bs_atime) ||
@@ -366,7 +391,7 @@ xfrog_bulkstat_v5_to_v1(
bs1->bs_blocks = bs5->bs_blocks;
bs1->bs_xflags = bs5->bs_xflags;
bs1->bs_extsize = cvt_off_fsb_to_b(xfd, bs5->bs_extsize_blks);
- bs1->bs_extents = bs5->bs_extents;
+ bs1->bs_extents = bs5->bs_extents64;
bs1->bs_gen = bs5->bs_gen;
bs1->bs_projid_lo = bs5->bs_projectid & 0xFFFF;
bs1->bs_forkoff = bs5->bs_forkoff;
@@ -407,7 +432,6 @@ xfrog_bulkstat_v1_to_v5(
bs5->bs_blocks = bs1->bs_blocks;
bs5->bs_xflags = bs1->bs_xflags;
bs5->bs_extsize_blks = cvt_b_to_off_fsbt(xfd, bs1->bs_extsize);
- bs5->bs_extents = bs1->bs_extents;
bs5->bs_gen = bs1->bs_gen;
bs5->bs_projectid = bstat_get_projid(bs1);
bs5->bs_forkoff = bs1->bs_forkoff;
@@ -415,6 +439,7 @@ xfrog_bulkstat_v1_to_v5(
bs5->bs_checked = bs1->bs_checked;
bs5->bs_cowextsize_blks = cvt_b_to_off_fsbt(xfd, bs1->bs_cowextsize);
bs5->bs_aextents = bs1->bs_aextents;
+ bs5->bs_extents64 = bs1->bs_extents;
}
/* Allocate a bulkstat request. Returns zero or a negative error code. */
diff --git a/libxfs/xfs_fs.h b/libxfs/xfs_fs.h
index 7260b140..6a9e51ac 100644
--- a/libxfs/xfs_fs.h
+++ b/libxfs/xfs_fs.h
@@ -391,7 +391,7 @@ struct xfs_bulkstat {
uint32_t bs_extsize_blks; /* extent size hint, blocks */
uint32_t bs_nlink; /* number of links */
- uint32_t bs_extents; /* number of extents */
+ uint32_t bs_extents; /* 32-bit data fork extent counter */
uint32_t bs_aextents; /* attribute number of extents */
uint16_t bs_version; /* structure version */
uint16_t bs_forkoff; /* inode fork offset in bytes */
@@ -400,8 +400,9 @@ struct xfs_bulkstat {
uint16_t bs_checked; /* checked inode metadata */
uint16_t bs_mode; /* type and mode */
uint16_t bs_pad2; /* zeroed */
+ uint64_t bs_extents64; /* 64-bit data fork extent counter */
- uint64_t bs_pad[7]; /* zeroed */
+ uint64_t bs_pad[6]; /* zeroed */
};
#define XFS_BULKSTAT_VERSION_V1 (1)
@@ -482,8 +483,11 @@ struct xfs_bulk_ireq {
*/
#define XFS_BULK_IREQ_SPECIAL (1 << 1)
-#define XFS_BULK_IREQ_FLAGS_ALL (XFS_BULK_IREQ_AGNO | \
- XFS_BULK_IREQ_SPECIAL)
+#define XFS_BULK_IREQ_NREXT64 (1 << 3)
+
+#define XFS_BULK_IREQ_FLAGS_ALL (XFS_BULK_IREQ_AGNO | \
+ XFS_BULK_IREQ_SPECIAL | \
+ XFS_BULK_IREQ_NREXT64)
/* Operate on the root directory inode. */
#define XFS_BULK_IREQ_SPECIAL_ROOT (1)
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH V4 15/20] xfsprogs: Enable bulkstat ioctl to support 64-bit extent counters
2021-12-14 8:48 ` [PATCH V4 15/20] xfsprogs: Enable bulkstat ioctl to support " Chandan Babu R
@ 2022-01-05 1:13 ` Darrick J. Wong
0 siblings, 0 replies; 34+ messages in thread
From: Darrick J. Wong @ 2022-01-05 1:13 UTC (permalink / raw)
To: Chandan Babu R; +Cc: linux-xfs, david
On Tue, Dec 14, 2021 at 02:18:06PM +0530, Chandan Babu R wrote:
> This commit adds support to libfrog to enable reporting 64-bit extent counters
> to its users. In order to do so, bulkstat ioctl is now invoked with the newly
> introduced XFS_BULK_IREQ_NREXT64 flag if the underlying filesystem's geometry
> supports 64-bit extent counters.
>
> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
For all the non libxfs/ changes,
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
--D
> ---
> fsr/xfs_fsr.c | 4 ++--
> io/bulkstat.c | 1 +
> libfrog/bulkstat.c | 29 +++++++++++++++++++++++++++--
> libxfs/xfs_fs.h | 12 ++++++++----
> 4 files changed, 38 insertions(+), 8 deletions(-)
>
> diff --git a/fsr/xfs_fsr.c b/fsr/xfs_fsr.c
> index 6cf8bfb7..ba02506d 100644
> --- a/fsr/xfs_fsr.c
> +++ b/fsr/xfs_fsr.c
> @@ -590,7 +590,7 @@ cmp(const void *s1, const void *s2)
> (bs1->bs_version == XFS_BULKSTAT_VERSION_V5 &&
> bs2->bs_version == XFS_BULKSTAT_VERSION_V5));
>
> - return (bs2->bs_extents - bs1->bs_extents);
> + return (bs2->bs_extents64 - bs1->bs_extents64);
> }
>
> /*
> @@ -655,7 +655,7 @@ fsrfs(char *mntdir, xfs_ino_t startino, int targetrange)
> for (p = buf, endp = (buf + buflenout); p < endp ; p++) {
> /* Do some obvious checks now */
> if (((p->bs_mode & S_IFMT) != S_IFREG) ||
> - (p->bs_extents < 2))
> + (p->bs_extents64 < 2))
> continue;
>
> ret = -xfrog_bulkstat_v5_to_v1(&fsxfd, &bs1, p);
> diff --git a/io/bulkstat.c b/io/bulkstat.c
> index 201470b2..0c9a2b02 100644
> --- a/io/bulkstat.c
> +++ b/io/bulkstat.c
> @@ -57,6 +57,7 @@ dump_bulkstat(
> printf("\tbs_sick = 0x%"PRIx16"\n", bstat->bs_sick);
> printf("\tbs_checked = 0x%"PRIx16"\n", bstat->bs_checked);
> printf("\tbs_mode = 0%"PRIo16"\n", bstat->bs_mode);
> + printf("\tbs_extents64 = %"PRIu64"\n", bstat->bs_extents64);
> };
>
> static void
> diff --git a/libfrog/bulkstat.c b/libfrog/bulkstat.c
> index 195f6ea0..0a90947f 100644
> --- a/libfrog/bulkstat.c
> +++ b/libfrog/bulkstat.c
> @@ -56,6 +56,9 @@ xfrog_bulkstat_single5(
> if (flags & ~(XFS_BULK_IREQ_SPECIAL))
> return -EINVAL;
>
> + if (xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64)
> + flags |= XFS_BULK_IREQ_NREXT64;
> +
> ret = xfrog_bulkstat_alloc_req(1, ino, &req);
> if (ret)
> return ret;
> @@ -73,6 +76,12 @@ xfrog_bulkstat_single5(
> }
>
> memcpy(bulkstat, req->bulkstat, sizeof(struct xfs_bulkstat));
> +
> + if (!(xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64)) {
> + bulkstat->bs_extents64 = bulkstat->bs_extents;
> + bulkstat->bs_extents = 0;
> + }
> +
> free:
> free(req);
> return ret;
> @@ -129,6 +138,7 @@ xfrog_bulkstat_single(
> switch (error) {
> case -EOPNOTSUPP:
> case -ENOTTY:
> + assert(!(xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64));
> xfd->flags |= XFROG_FLAG_BULKSTAT_FORCE_V1;
> break;
> }
> @@ -259,10 +269,23 @@ xfrog_bulkstat5(
> struct xfs_bulkstat_req *req)
> {
> int ret;
> + int i;
> +
> + if (xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64)
> + req->hdr.flags |= XFS_BULK_IREQ_NREXT64;
>
> ret = ioctl(xfd->fd, XFS_IOC_BULKSTAT, req);
> if (ret)
> return -errno;
> +
> + if (!(xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64)) {
> + for (i = 0; i < req->hdr.ocount; i++) {
> + req->bulkstat[i].bs_extents64 =
> + req->bulkstat[i].bs_extents;
> + req->bulkstat[i].bs_extents = 0;
> + }
> + }
> +
> return 0;
> }
>
> @@ -316,6 +339,7 @@ xfrog_bulkstat(
> switch (error) {
> case -EOPNOTSUPP:
> case -ENOTTY:
> + assert(!(xfd->fsgeom.flags & XFS_FSOP_GEOM_FLAGS_NREXT64));
> xfd->flags |= XFROG_FLAG_BULKSTAT_FORCE_V1;
> break;
> }
> @@ -342,6 +366,7 @@ xfrog_bulkstat_v5_to_v1(
> const struct xfs_bulkstat *bs5)
> {
> if (bs5->bs_aextents > UINT16_MAX ||
> + bs5->bs_extents64 > INT32_MAX ||
> cvt_off_fsb_to_b(xfd, bs5->bs_extsize_blks) > UINT32_MAX ||
> cvt_off_fsb_to_b(xfd, bs5->bs_cowextsize_blks) > UINT32_MAX ||
> time_too_big(bs5->bs_atime) ||
> @@ -366,7 +391,7 @@ xfrog_bulkstat_v5_to_v1(
> bs1->bs_blocks = bs5->bs_blocks;
> bs1->bs_xflags = bs5->bs_xflags;
> bs1->bs_extsize = cvt_off_fsb_to_b(xfd, bs5->bs_extsize_blks);
> - bs1->bs_extents = bs5->bs_extents;
> + bs1->bs_extents = bs5->bs_extents64;
> bs1->bs_gen = bs5->bs_gen;
> bs1->bs_projid_lo = bs5->bs_projectid & 0xFFFF;
> bs1->bs_forkoff = bs5->bs_forkoff;
> @@ -407,7 +432,6 @@ xfrog_bulkstat_v1_to_v5(
> bs5->bs_blocks = bs1->bs_blocks;
> bs5->bs_xflags = bs1->bs_xflags;
> bs5->bs_extsize_blks = cvt_b_to_off_fsbt(xfd, bs1->bs_extsize);
> - bs5->bs_extents = bs1->bs_extents;
> bs5->bs_gen = bs1->bs_gen;
> bs5->bs_projectid = bstat_get_projid(bs1);
> bs5->bs_forkoff = bs1->bs_forkoff;
> @@ -415,6 +439,7 @@ xfrog_bulkstat_v1_to_v5(
> bs5->bs_checked = bs1->bs_checked;
> bs5->bs_cowextsize_blks = cvt_b_to_off_fsbt(xfd, bs1->bs_cowextsize);
> bs5->bs_aextents = bs1->bs_aextents;
> + bs5->bs_extents64 = bs1->bs_extents;
> }
>
> /* Allocate a bulkstat request. Returns zero or a negative error code. */
> diff --git a/libxfs/xfs_fs.h b/libxfs/xfs_fs.h
> index 7260b140..6a9e51ac 100644
> --- a/libxfs/xfs_fs.h
> +++ b/libxfs/xfs_fs.h
> @@ -391,7 +391,7 @@ struct xfs_bulkstat {
> uint32_t bs_extsize_blks; /* extent size hint, blocks */
>
> uint32_t bs_nlink; /* number of links */
> - uint32_t bs_extents; /* number of extents */
> + uint32_t bs_extents; /* 32-bit data fork extent counter */
> uint32_t bs_aextents; /* attribute number of extents */
> uint16_t bs_version; /* structure version */
> uint16_t bs_forkoff; /* inode fork offset in bytes */
> @@ -400,8 +400,9 @@ struct xfs_bulkstat {
> uint16_t bs_checked; /* checked inode metadata */
> uint16_t bs_mode; /* type and mode */
> uint16_t bs_pad2; /* zeroed */
> + uint64_t bs_extents64; /* 64-bit data fork extent counter */
>
> - uint64_t bs_pad[7]; /* zeroed */
> + uint64_t bs_pad[6]; /* zeroed */
> };
>
> #define XFS_BULKSTAT_VERSION_V1 (1)
> @@ -482,8 +483,11 @@ struct xfs_bulk_ireq {
> */
> #define XFS_BULK_IREQ_SPECIAL (1 << 1)
>
> -#define XFS_BULK_IREQ_FLAGS_ALL (XFS_BULK_IREQ_AGNO | \
> - XFS_BULK_IREQ_SPECIAL)
> +#define XFS_BULK_IREQ_NREXT64 (1 << 3)
> +
> +#define XFS_BULK_IREQ_FLAGS_ALL (XFS_BULK_IREQ_AGNO | \
> + XFS_BULK_IREQ_SPECIAL | \
> + XFS_BULK_IREQ_NREXT64)
>
> /* Operate on the root directory inode. */
> #define XFS_BULK_IREQ_SPECIAL_ROOT (1)
> --
> 2.30.2
>
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH V4 16/20] xfsprogs: Add XFS_SB_FEAT_INCOMPAT_NREXT64 to the list of supported flags
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (14 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 15/20] xfsprogs: Enable bulkstat ioctl to support " Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2021-12-14 8:48 ` [PATCH V4 17/20] xfsprogs: xfs_info: Report NREXT64 feature status Chandan Babu R
` (3 subsequent siblings)
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
This commit enables XFS module to work with fs instances having 64-bit
per-inode extent counters by adding XFS_SB_FEAT_INCOMPAT_NREXT64 flag to the
list of supported incompat feature flags.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libxfs/xfs_format.h | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index d4c962fc..0ed9d5ac 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -475,7 +475,8 @@ xfs_sb_has_ro_compat_feature(
XFS_SB_FEAT_INCOMPAT_SPINODES| \
XFS_SB_FEAT_INCOMPAT_META_UUID| \
XFS_SB_FEAT_INCOMPAT_BIGTIME| \
- XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR)
+ XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR| \
+ XFS_SB_FEAT_INCOMPAT_NREXT64)
#define XFS_SB_FEAT_INCOMPAT_UNKNOWN ~XFS_SB_FEAT_INCOMPAT_ALL
static inline bool
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH V4 17/20] xfsprogs: xfs_info: Report NREXT64 feature status
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (15 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 16/20] xfsprogs: Add XFS_SB_FEAT_INCOMPAT_NREXT64 to the list of supported flags Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2022-01-05 1:13 ` Darrick J. Wong
2021-12-14 8:48 ` [PATCH V4 18/20] xfsprogs: Add mkfs option to create filesystem with large extent counters Chandan Babu R
` (2 subsequent siblings)
19 siblings, 1 reply; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
This commit adds support to libfrog to obtain information about the
availability of NREXT64 feature in the underlying filesystem.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
libfrog/fsgeom.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/libfrog/fsgeom.c b/libfrog/fsgeom.c
index 4f1a1842..3e7f0797 100644
--- a/libfrog/fsgeom.c
+++ b/libfrog/fsgeom.c
@@ -30,6 +30,7 @@ xfs_report_geom(
int reflink_enabled;
int bigtime_enabled;
int inobtcount;
+ int nrext64;
isint = geo->logstart > 0;
lazycount = geo->flags & XFS_FSOP_GEOM_FLAGS_LAZYSB ? 1 : 0;
@@ -47,12 +48,13 @@ xfs_report_geom(
reflink_enabled = geo->flags & XFS_FSOP_GEOM_FLAGS_REFLINK ? 1 : 0;
bigtime_enabled = geo->flags & XFS_FSOP_GEOM_FLAGS_BIGTIME ? 1 : 0;
inobtcount = geo->flags & XFS_FSOP_GEOM_FLAGS_INOBTCNT ? 1 : 0;
+ nrext64 = geo->flags & XFS_FSOP_GEOM_FLAGS_NREXT64 ? 1 : 0;
printf(_(
"meta-data=%-22s isize=%-6d agcount=%u, agsize=%u blks\n"
" =%-22s sectsz=%-5u attr=%u, projid32bit=%u\n"
" =%-22s crc=%-8u finobt=%u, sparse=%u, rmapbt=%u\n"
-" =%-22s reflink=%-4u bigtime=%u inobtcount=%u\n"
+" =%-22s reflink=%-4u bigtime=%u inobtcount=%u nrext64=%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"
@@ -62,7 +64,7 @@ xfs_report_geom(
mntpoint, geo->inodesize, geo->agcount, geo->agblocks,
"", geo->sectsize, attrversion, projid32bit,
"", crcs_enabled, finobt_enabled, spinodes, rmapbt_enabled,
- "", reflink_enabled, bigtime_enabled, inobtcount,
+ "", reflink_enabled, bigtime_enabled, inobtcount, nrext64,
"", geo->blocksize, (unsigned long long)geo->datablocks,
geo->imaxpct,
"", geo->sunit, geo->swidth,
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH V4 17/20] xfsprogs: xfs_info: Report NREXT64 feature status
2021-12-14 8:48 ` [PATCH V4 17/20] xfsprogs: xfs_info: Report NREXT64 feature status Chandan Babu R
@ 2022-01-05 1:13 ` Darrick J. Wong
0 siblings, 0 replies; 34+ messages in thread
From: Darrick J. Wong @ 2022-01-05 1:13 UTC (permalink / raw)
To: Chandan Babu R; +Cc: linux-xfs, david
On Tue, Dec 14, 2021 at 02:18:08PM +0530, Chandan Babu R wrote:
> This commit adds support to libfrog to obtain information about the
> availability of NREXT64 feature in the underlying filesystem.
>
> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
LGTM
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
--D
> ---
> libfrog/fsgeom.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/libfrog/fsgeom.c b/libfrog/fsgeom.c
> index 4f1a1842..3e7f0797 100644
> --- a/libfrog/fsgeom.c
> +++ b/libfrog/fsgeom.c
> @@ -30,6 +30,7 @@ xfs_report_geom(
> int reflink_enabled;
> int bigtime_enabled;
> int inobtcount;
> + int nrext64;
>
> isint = geo->logstart > 0;
> lazycount = geo->flags & XFS_FSOP_GEOM_FLAGS_LAZYSB ? 1 : 0;
> @@ -47,12 +48,13 @@ xfs_report_geom(
> reflink_enabled = geo->flags & XFS_FSOP_GEOM_FLAGS_REFLINK ? 1 : 0;
> bigtime_enabled = geo->flags & XFS_FSOP_GEOM_FLAGS_BIGTIME ? 1 : 0;
> inobtcount = geo->flags & XFS_FSOP_GEOM_FLAGS_INOBTCNT ? 1 : 0;
> + nrext64 = geo->flags & XFS_FSOP_GEOM_FLAGS_NREXT64 ? 1 : 0;
>
> printf(_(
> "meta-data=%-22s isize=%-6d agcount=%u, agsize=%u blks\n"
> " =%-22s sectsz=%-5u attr=%u, projid32bit=%u\n"
> " =%-22s crc=%-8u finobt=%u, sparse=%u, rmapbt=%u\n"
> -" =%-22s reflink=%-4u bigtime=%u inobtcount=%u\n"
> +" =%-22s reflink=%-4u bigtime=%u inobtcount=%u nrext64=%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"
> @@ -62,7 +64,7 @@ xfs_report_geom(
> mntpoint, geo->inodesize, geo->agcount, geo->agblocks,
> "", geo->sectsize, attrversion, projid32bit,
> "", crcs_enabled, finobt_enabled, spinodes, rmapbt_enabled,
> - "", reflink_enabled, bigtime_enabled, inobtcount,
> + "", reflink_enabled, bigtime_enabled, inobtcount, nrext64,
> "", geo->blocksize, (unsigned long long)geo->datablocks,
> geo->imaxpct,
> "", geo->sunit, geo->swidth,
> --
> 2.30.2
>
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH V4 18/20] xfsprogs: Add mkfs option to create filesystem with large extent counters
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (16 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 17/20] xfsprogs: xfs_info: Report NREXT64 feature status Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2022-01-05 1:15 ` Darrick J. Wong
2021-12-14 8:48 ` [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature Chandan Babu R
2021-12-14 8:48 ` [PATCH V4 20/20] xfsprogs: Define max extent length based on on-disk format definition Chandan Babu R
19 siblings, 1 reply; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
Enabling nrext64 option on mkfs.xfs command line extends the maximum values of
inode data and attr fork extent counters to 2^48 - 1 and 2^32 - 1
respectively. This also sets the XFS_SB_FEAT_INCOMPAT_NREXT64 incompat flag
on the superblock preventing older kernels from mounting such a filesystem.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
man/man8/mkfs.xfs.8 | 7 +++++++
mkfs/xfs_mkfs.c | 23 +++++++++++++++++++++++
2 files changed, 30 insertions(+)
diff --git a/man/man8/mkfs.xfs.8 b/man/man8/mkfs.xfs.8
index a7f70285..923940e3 100644
--- a/man/man8/mkfs.xfs.8
+++ b/man/man8/mkfs.xfs.8
@@ -639,6 +639,13 @@ space over time such that no free extents are large enough to
accommodate a chunk of 64 inodes. Without this feature enabled, inode
allocations can fail with out of space errors under severe fragmented
free space conditions.
+.TP
+.BI nrext64[= value]
+Extend maximum values of inode data and attr fork extent counters from 2^31 -
+1 and 2^15 - 1 to 2^48 - 1 and 2^32 - 1 respectively. If the value is
+omitted, 1 is assumed. This feature is disabled by default. This feature is
+only available for filesystems formatted with -m crc=1.
+.TP
.RE
.PP
.PD 0
diff --git a/mkfs/xfs_mkfs.c b/mkfs/xfs_mkfs.c
index 53904677..6609776f 100644
--- a/mkfs/xfs_mkfs.c
+++ b/mkfs/xfs_mkfs.c
@@ -78,6 +78,7 @@ enum {
I_ATTR,
I_PROJID32BIT,
I_SPINODES,
+ I_NREXT64,
I_MAX_OPTS,
};
@@ -432,6 +433,7 @@ static struct opt_params iopts = {
[I_ATTR] = "attr",
[I_PROJID32BIT] = "projid32bit",
[I_SPINODES] = "sparse",
+ [I_NREXT64] = "nrext64",
},
.subopt_params = {
{ .index = I_ALIGN,
@@ -480,6 +482,12 @@ static struct opt_params iopts = {
.maxval = 1,
.defaultval = 1,
},
+ { .index = I_NREXT64,
+ .conflicts = { { NULL, LAST_CONFLICT } },
+ .minval = 0,
+ .maxval = 1,
+ .defaultval = 1,
+ }
},
};
@@ -804,6 +812,7 @@ struct sb_feat_args {
bool bigtime; /* XFS_SB_FEAT_INCOMPAT_BIGTIME */
bool nodalign;
bool nortalign;
+ bool nrext64;
};
struct cli_params {
@@ -1585,6 +1594,9 @@ inode_opts_parser(
case I_SPINODES:
cli->sb_feat.spinodes = getnum(value, opts, subopt);
break;
+ case I_NREXT64:
+ cli->sb_feat.nrext64 = getnum(value, opts, subopt);
+ break;
default:
return -EINVAL;
}
@@ -2153,6 +2165,14 @@ _("timestamps later than 2038 not supported without CRC support\n"));
usage();
}
cli->sb_feat.bigtime = false;
+
+ if (cli->sb_feat.nrext64 &&
+ cli_opt_set(&iopts, I_NREXT64)) {
+ fprintf(stderr,
+_("64 bit extent count not supported without CRC support\n"));
+ usage();
+ }
+ cli->sb_feat.nrext64 = false;
}
if (!cli->sb_feat.finobt) {
@@ -3145,6 +3165,8 @@ sb_set_features(
sbp->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_SPINODES;
}
+ if (fp->nrext64)
+ sbp->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NREXT64;
}
/*
@@ -3854,6 +3876,7 @@ main(
.nodalign = false,
.nortalign = false,
.bigtime = false,
+ .nrext64 = false,
},
};
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH V4 18/20] xfsprogs: Add mkfs option to create filesystem with large extent counters
2021-12-14 8:48 ` [PATCH V4 18/20] xfsprogs: Add mkfs option to create filesystem with large extent counters Chandan Babu R
@ 2022-01-05 1:15 ` Darrick J. Wong
0 siblings, 0 replies; 34+ messages in thread
From: Darrick J. Wong @ 2022-01-05 1:15 UTC (permalink / raw)
To: Chandan Babu R; +Cc: linux-xfs, david
On Tue, Dec 14, 2021 at 02:18:09PM +0530, Chandan Babu R wrote:
> Enabling nrext64 option on mkfs.xfs command line extends the maximum values of
> inode data and attr fork extent counters to 2^48 - 1 and 2^32 - 1
> respectively. This also sets the XFS_SB_FEAT_INCOMPAT_NREXT64 incompat flag
> on the superblock preventing older kernels from mounting such a filesystem.
>
> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
Looks reasonable. There's a silly nit that mtk prefers that each
sentence in a Linux man page start on a separate line, but I'll let Eric
decide if he cares about that for xfsprogs manpages.
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
--D
> ---
> man/man8/mkfs.xfs.8 | 7 +++++++
> mkfs/xfs_mkfs.c | 23 +++++++++++++++++++++++
> 2 files changed, 30 insertions(+)
>
> diff --git a/man/man8/mkfs.xfs.8 b/man/man8/mkfs.xfs.8
> index a7f70285..923940e3 100644
> --- a/man/man8/mkfs.xfs.8
> +++ b/man/man8/mkfs.xfs.8
> @@ -639,6 +639,13 @@ space over time such that no free extents are large enough to
> accommodate a chunk of 64 inodes. Without this feature enabled, inode
> allocations can fail with out of space errors under severe fragmented
> free space conditions.
> +.TP
> +.BI nrext64[= value]
> +Extend maximum values of inode data and attr fork extent counters from 2^31 -
> +1 and 2^15 - 1 to 2^48 - 1 and 2^32 - 1 respectively. If the value is
> +omitted, 1 is assumed. This feature is disabled by default. This feature is
> +only available for filesystems formatted with -m crc=1.
> +.TP
> .RE
> .PP
> .PD 0
> diff --git a/mkfs/xfs_mkfs.c b/mkfs/xfs_mkfs.c
> index 53904677..6609776f 100644
> --- a/mkfs/xfs_mkfs.c
> +++ b/mkfs/xfs_mkfs.c
> @@ -78,6 +78,7 @@ enum {
> I_ATTR,
> I_PROJID32BIT,
> I_SPINODES,
> + I_NREXT64,
> I_MAX_OPTS,
> };
>
> @@ -432,6 +433,7 @@ static struct opt_params iopts = {
> [I_ATTR] = "attr",
> [I_PROJID32BIT] = "projid32bit",
> [I_SPINODES] = "sparse",
> + [I_NREXT64] = "nrext64",
> },
> .subopt_params = {
> { .index = I_ALIGN,
> @@ -480,6 +482,12 @@ static struct opt_params iopts = {
> .maxval = 1,
> .defaultval = 1,
> },
> + { .index = I_NREXT64,
> + .conflicts = { { NULL, LAST_CONFLICT } },
> + .minval = 0,
> + .maxval = 1,
> + .defaultval = 1,
> + }
> },
> };
>
> @@ -804,6 +812,7 @@ struct sb_feat_args {
> bool bigtime; /* XFS_SB_FEAT_INCOMPAT_BIGTIME */
> bool nodalign;
> bool nortalign;
> + bool nrext64;
> };
>
> struct cli_params {
> @@ -1585,6 +1594,9 @@ inode_opts_parser(
> case I_SPINODES:
> cli->sb_feat.spinodes = getnum(value, opts, subopt);
> break;
> + case I_NREXT64:
> + cli->sb_feat.nrext64 = getnum(value, opts, subopt);
> + break;
> default:
> return -EINVAL;
> }
> @@ -2153,6 +2165,14 @@ _("timestamps later than 2038 not supported without CRC support\n"));
> usage();
> }
> cli->sb_feat.bigtime = false;
> +
> + if (cli->sb_feat.nrext64 &&
> + cli_opt_set(&iopts, I_NREXT64)) {
> + fprintf(stderr,
> +_("64 bit extent count not supported without CRC support\n"));
> + usage();
> + }
> + cli->sb_feat.nrext64 = false;
> }
>
> if (!cli->sb_feat.finobt) {
> @@ -3145,6 +3165,8 @@ sb_set_features(
> sbp->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_SPINODES;
> }
>
> + if (fp->nrext64)
> + sbp->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NREXT64;
> }
>
> /*
> @@ -3854,6 +3876,7 @@ main(
> .nodalign = false,
> .nortalign = false,
> .bigtime = false,
> + .nrext64 = false,
> },
> };
>
> --
> 2.30.2
>
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (17 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 18/20] xfsprogs: Add mkfs option to create filesystem with large extent counters Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
2022-01-05 1:17 ` Darrick J. Wong
2021-12-14 8:48 ` [PATCH V4 20/20] xfsprogs: Define max extent length based on on-disk format definition Chandan Babu R
19 siblings, 1 reply; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
This commit adds support to xfs_repair to allow upgrading an existing
filesystem to support per-inode large extent counters.
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
repair/globals.c | 1 +
repair/globals.h | 1 +
repair/phase2.c | 35 ++++++++++++++++++++++++++++++++++-
repair/xfs_repair.c | 11 +++++++++++
4 files changed, 47 insertions(+), 1 deletion(-)
diff --git a/repair/globals.c b/repair/globals.c
index d89507b1..2f29391a 100644
--- a/repair/globals.c
+++ b/repair/globals.c
@@ -53,6 +53,7 @@ bool add_bigtime; /* add support for timestamps up to 2486 */
bool add_finobt; /* add free inode btrees */
bool add_reflink; /* add reference count btrees */
bool add_rmapbt; /* add reverse mapping btrees */
+bool add_nrext64;
/* misc status variables */
diff --git a/repair/globals.h b/repair/globals.h
index 53ff2532..af0bcb6b 100644
--- a/repair/globals.h
+++ b/repair/globals.h
@@ -94,6 +94,7 @@ extern bool add_bigtime; /* add support for timestamps up to 2486 */
extern bool add_finobt; /* add free inode btrees */
extern bool add_reflink; /* add reference count btrees */
extern bool add_rmapbt; /* add reverse mapping btrees */
+extern bool add_nrext64;
/* misc status variables */
diff --git a/repair/phase2.c b/repair/phase2.c
index c811ed5d..c9db3281 100644
--- a/repair/phase2.c
+++ b/repair/phase2.c
@@ -191,6 +191,7 @@ check_new_v5_geometry(
struct xfs_perag *pag;
xfs_agnumber_t agno;
xfs_ino_t rootino;
+ uint old_bm_maxlevels[2];
int min_logblocks;
int error;
@@ -201,6 +202,12 @@ check_new_v5_geometry(
memcpy(&old_sb, &mp->m_sb, sizeof(struct xfs_sb));
memcpy(&mp->m_sb, new_sb, sizeof(struct xfs_sb));
+ old_bm_maxlevels[0] = mp->m_bm_maxlevels[0];
+ old_bm_maxlevels[1] = mp->m_bm_maxlevels[1];
+
+ xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
+ xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK);
+
/* Do we have a big enough log? */
min_logblocks = libxfs_log_calc_minimum_size(mp);
if (old_sb.sb_logblocks < min_logblocks) {
@@ -288,6 +295,9 @@ check_new_v5_geometry(
pag->pagi_init = 0;
}
+ mp->m_bm_maxlevels[0] = old_bm_maxlevels[0];
+ mp->m_bm_maxlevels[1] = old_bm_maxlevels[1];
+
/*
* Put back the old superblock.
*/
@@ -366,6 +376,28 @@ set_rmapbt(
return true;
}
+static bool
+set_nrext64(
+ struct xfs_mount *mp,
+ struct xfs_sb *new_sb)
+{
+ if (!xfs_sb_version_hascrc(&mp->m_sb)) {
+ printf(
+ _("Nrext64 only supported on V5 filesystems.\n"));
+ exit(0);
+ }
+
+ if (xfs_sb_version_hasnrext64(&mp->m_sb)) {
+ printf(_("Filesystem already supports nrext64.\n"));
+ exit(0);
+ }
+
+ printf(_("Adding nrext64 to filesystem.\n"));
+ new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NREXT64;
+ new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
+ return true;
+}
+
/* Perform the user's requested upgrades on filesystem. */
static void
upgrade_filesystem(
@@ -388,7 +420,8 @@ upgrade_filesystem(
dirty |= set_reflink(mp, &new_sb);
if (add_rmapbt)
dirty |= set_rmapbt(mp, &new_sb);
-
+ if (add_nrext64)
+ dirty |= set_nrext64(mp, &new_sb);
if (!dirty)
return;
diff --git a/repair/xfs_repair.c b/repair/xfs_repair.c
index e250a5bf..96c9bb56 100644
--- a/repair/xfs_repair.c
+++ b/repair/xfs_repair.c
@@ -70,6 +70,7 @@ enum c_opt_nums {
CONVERT_FINOBT,
CONVERT_REFLINK,
CONVERT_RMAPBT,
+ CONVERT_NREXT64,
C_MAX_OPTS,
};
@@ -80,6 +81,7 @@ static char *c_opts[] = {
[CONVERT_FINOBT] = "finobt",
[CONVERT_REFLINK] = "reflink",
[CONVERT_RMAPBT] = "rmapbt",
+ [CONVERT_NREXT64] = "nrext64",
[C_MAX_OPTS] = NULL,
};
@@ -357,6 +359,15 @@ process_args(int argc, char **argv)
_("-c rmapbt only supports upgrades\n"));
add_rmapbt = true;
break;
+ case CONVERT_NREXT64:
+ if (!val)
+ do_abort(
+ _("-c nrext64 requires a parameter\n"));
+ if (strtol(val, NULL, 0) != 1)
+ do_abort(
+ _("-c nrext64 only supports upgrades\n"));
+ add_nrext64 = true;
+ break;
default:
unknown('c', val);
break;
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature
2021-12-14 8:48 ` [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature Chandan Babu R
@ 2022-01-05 1:17 ` Darrick J. Wong
2022-01-07 16:17 ` Chandan Babu R
0 siblings, 1 reply; 34+ messages in thread
From: Darrick J. Wong @ 2022-01-05 1:17 UTC (permalink / raw)
To: Chandan Babu R; +Cc: linux-xfs, david
On Tue, Dec 14, 2021 at 02:18:10PM +0530, Chandan Babu R wrote:
> This commit adds support to xfs_repair to allow upgrading an existing
> filesystem to support per-inode large extent counters.
>
> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
> ---
> repair/globals.c | 1 +
> repair/globals.h | 1 +
> repair/phase2.c | 35 ++++++++++++++++++++++++++++++++++-
> repair/xfs_repair.c | 11 +++++++++++
> 4 files changed, 47 insertions(+), 1 deletion(-)
>
> diff --git a/repair/globals.c b/repair/globals.c
> index d89507b1..2f29391a 100644
> --- a/repair/globals.c
> +++ b/repair/globals.c
> @@ -53,6 +53,7 @@ bool add_bigtime; /* add support for timestamps up to 2486 */
> bool add_finobt; /* add free inode btrees */
> bool add_reflink; /* add reference count btrees */
> bool add_rmapbt; /* add reverse mapping btrees */
> +bool add_nrext64;
>
> /* misc status variables */
>
> diff --git a/repair/globals.h b/repair/globals.h
> index 53ff2532..af0bcb6b 100644
> --- a/repair/globals.h
> +++ b/repair/globals.h
> @@ -94,6 +94,7 @@ extern bool add_bigtime; /* add support for timestamps up to 2486 */
> extern bool add_finobt; /* add free inode btrees */
> extern bool add_reflink; /* add reference count btrees */
> extern bool add_rmapbt; /* add reverse mapping btrees */
> +extern bool add_nrext64;
>
> /* misc status variables */
>
> diff --git a/repair/phase2.c b/repair/phase2.c
> index c811ed5d..c9db3281 100644
> --- a/repair/phase2.c
> +++ b/repair/phase2.c
> @@ -191,6 +191,7 @@ check_new_v5_geometry(
> struct xfs_perag *pag;
> xfs_agnumber_t agno;
> xfs_ino_t rootino;
> + uint old_bm_maxlevels[2];
> int min_logblocks;
> int error;
>
> @@ -201,6 +202,12 @@ check_new_v5_geometry(
> memcpy(&old_sb, &mp->m_sb, sizeof(struct xfs_sb));
> memcpy(&mp->m_sb, new_sb, sizeof(struct xfs_sb));
>
> + old_bm_maxlevels[0] = mp->m_bm_maxlevels[0];
> + old_bm_maxlevels[1] = mp->m_bm_maxlevels[1];
> +
> + xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
> + xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK);
Ahh... I see why you added my (evil) patch that allows upgrading a
filesystem to reflink -- you need the check_new_v5_geometry function so
that you can check if the log size is big enough to handle larger bmbt
trees.
Hmm, I guess I should work on separating this from the actual
rmap/reflink/finobt upgrade code, since I have no idea if we /ever/ want
to support that.
--D
> +
> /* Do we have a big enough log? */
> min_logblocks = libxfs_log_calc_minimum_size(mp);
> if (old_sb.sb_logblocks < min_logblocks) {
> @@ -288,6 +295,9 @@ check_new_v5_geometry(
> pag->pagi_init = 0;
> }
>
> + mp->m_bm_maxlevels[0] = old_bm_maxlevels[0];
> + mp->m_bm_maxlevels[1] = old_bm_maxlevels[1];
> +
> /*
> * Put back the old superblock.
> */
> @@ -366,6 +376,28 @@ set_rmapbt(
> return true;
> }
>
> +static bool
> +set_nrext64(
> + struct xfs_mount *mp,
> + struct xfs_sb *new_sb)
> +{
> + if (!xfs_sb_version_hascrc(&mp->m_sb)) {
> + printf(
> + _("Nrext64 only supported on V5 filesystems.\n"));
> + exit(0);
> + }
> +
> + if (xfs_sb_version_hasnrext64(&mp->m_sb)) {
> + printf(_("Filesystem already supports nrext64.\n"));
> + exit(0);
> + }
> +
> + printf(_("Adding nrext64 to filesystem.\n"));
> + new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NREXT64;
> + new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
> + return true;
> +}
> +
> /* Perform the user's requested upgrades on filesystem. */
> static void
> upgrade_filesystem(
> @@ -388,7 +420,8 @@ upgrade_filesystem(
> dirty |= set_reflink(mp, &new_sb);
> if (add_rmapbt)
> dirty |= set_rmapbt(mp, &new_sb);
> -
> + if (add_nrext64)
> + dirty |= set_nrext64(mp, &new_sb);
> if (!dirty)
> return;
>
> diff --git a/repair/xfs_repair.c b/repair/xfs_repair.c
> index e250a5bf..96c9bb56 100644
> --- a/repair/xfs_repair.c
> +++ b/repair/xfs_repair.c
> @@ -70,6 +70,7 @@ enum c_opt_nums {
> CONVERT_FINOBT,
> CONVERT_REFLINK,
> CONVERT_RMAPBT,
> + CONVERT_NREXT64,
> C_MAX_OPTS,
> };
>
> @@ -80,6 +81,7 @@ static char *c_opts[] = {
> [CONVERT_FINOBT] = "finobt",
> [CONVERT_REFLINK] = "reflink",
> [CONVERT_RMAPBT] = "rmapbt",
> + [CONVERT_NREXT64] = "nrext64",
> [C_MAX_OPTS] = NULL,
> };
>
> @@ -357,6 +359,15 @@ process_args(int argc, char **argv)
> _("-c rmapbt only supports upgrades\n"));
> add_rmapbt = true;
> break;
> + case CONVERT_NREXT64:
> + if (!val)
> + do_abort(
> + _("-c nrext64 requires a parameter\n"));
> + if (strtol(val, NULL, 0) != 1)
> + do_abort(
> + _("-c nrext64 only supports upgrades\n"));
> + add_nrext64 = true;
> + break;
> default:
> unknown('c', val);
> break;
> --
> 2.30.2
>
^ permalink raw reply [flat|nested] 34+ messages in thread* Re: [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature
2022-01-05 1:17 ` Darrick J. Wong
@ 2022-01-07 16:17 ` Chandan Babu R
2022-01-07 19:03 ` Darrick J. Wong
0 siblings, 1 reply; 34+ messages in thread
From: Chandan Babu R @ 2022-01-07 16:17 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: linux-xfs, david
On 05 Jan 2022 at 06:47, Darrick J. Wong wrote:
> On Tue, Dec 14, 2021 at 02:18:10PM +0530, Chandan Babu R wrote:
>> This commit adds support to xfs_repair to allow upgrading an existing
>> filesystem to support per-inode large extent counters.
>>
>> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
>> ---
>> repair/globals.c | 1 +
>> repair/globals.h | 1 +
>> repair/phase2.c | 35 ++++++++++++++++++++++++++++++++++-
>> repair/xfs_repair.c | 11 +++++++++++
>> 4 files changed, 47 insertions(+), 1 deletion(-)
>>
>> diff --git a/repair/globals.c b/repair/globals.c
>> index d89507b1..2f29391a 100644
>> --- a/repair/globals.c
>> +++ b/repair/globals.c
>> @@ -53,6 +53,7 @@ bool add_bigtime; /* add support for timestamps up to 2486 */
>> bool add_finobt; /* add free inode btrees */
>> bool add_reflink; /* add reference count btrees */
>> bool add_rmapbt; /* add reverse mapping btrees */
>> +bool add_nrext64;
>>
>> /* misc status variables */
>>
>> diff --git a/repair/globals.h b/repair/globals.h
>> index 53ff2532..af0bcb6b 100644
>> --- a/repair/globals.h
>> +++ b/repair/globals.h
>> @@ -94,6 +94,7 @@ extern bool add_bigtime; /* add support for timestamps up to 2486 */
>> extern bool add_finobt; /* add free inode btrees */
>> extern bool add_reflink; /* add reference count btrees */
>> extern bool add_rmapbt; /* add reverse mapping btrees */
>> +extern bool add_nrext64;
>>
>> /* misc status variables */
>>
>> diff --git a/repair/phase2.c b/repair/phase2.c
>> index c811ed5d..c9db3281 100644
>> --- a/repair/phase2.c
>> +++ b/repair/phase2.c
>> @@ -191,6 +191,7 @@ check_new_v5_geometry(
>> struct xfs_perag *pag;
>> xfs_agnumber_t agno;
>> xfs_ino_t rootino;
>> + uint old_bm_maxlevels[2];
>> int min_logblocks;
>> int error;
>>
>> @@ -201,6 +202,12 @@ check_new_v5_geometry(
>> memcpy(&old_sb, &mp->m_sb, sizeof(struct xfs_sb));
>> memcpy(&mp->m_sb, new_sb, sizeof(struct xfs_sb));
>>
>> + old_bm_maxlevels[0] = mp->m_bm_maxlevels[0];
>> + old_bm_maxlevels[1] = mp->m_bm_maxlevels[1];
>> +
>> + xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
>> + xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK);
>
> Ahh... I see why you added my (evil) patch that allows upgrading a
> filesystem to reflink -- you need the check_new_v5_geometry function so
> that you can check if the log size is big enough to handle larger bmbt
> trees.
>
> Hmm, I guess I should work on separating this from the actual
> rmap/reflink/finobt upgrade code, since I have no idea if we /ever/ want
> to support that.
>
I can do that. I will include the trimmed down version of the patch before
posting the patchset once again.
> --D
>
>> +
>> /* Do we have a big enough log? */
>> min_logblocks = libxfs_log_calc_minimum_size(mp);
>> if (old_sb.sb_logblocks < min_logblocks) {
>> @@ -288,6 +295,9 @@ check_new_v5_geometry(
>> pag->pagi_init = 0;
>> }
>>
>> + mp->m_bm_maxlevels[0] = old_bm_maxlevels[0];
>> + mp->m_bm_maxlevels[1] = old_bm_maxlevels[1];
>> +
>> /*
>> * Put back the old superblock.
>> */
>> @@ -366,6 +376,28 @@ set_rmapbt(
>> return true;
>> }
>>
>> +static bool
>> +set_nrext64(
>> + struct xfs_mount *mp,
>> + struct xfs_sb *new_sb)
>> +{
>> + if (!xfs_sb_version_hascrc(&mp->m_sb)) {
>> + printf(
>> + _("Nrext64 only supported on V5 filesystems.\n"));
>> + exit(0);
>> + }
>> +
>> + if (xfs_sb_version_hasnrext64(&mp->m_sb)) {
>> + printf(_("Filesystem already supports nrext64.\n"));
>> + exit(0);
>> + }
>> +
>> + printf(_("Adding nrext64 to filesystem.\n"));
>> + new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NREXT64;
>> + new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
>> + return true;
>> +}
>> +
>> /* Perform the user's requested upgrades on filesystem. */
>> static void
>> upgrade_filesystem(
>> @@ -388,7 +420,8 @@ upgrade_filesystem(
>> dirty |= set_reflink(mp, &new_sb);
>> if (add_rmapbt)
>> dirty |= set_rmapbt(mp, &new_sb);
>> -
>> + if (add_nrext64)
>> + dirty |= set_nrext64(mp, &new_sb);
>> if (!dirty)
>> return;
>>
>> diff --git a/repair/xfs_repair.c b/repair/xfs_repair.c
>> index e250a5bf..96c9bb56 100644
>> --- a/repair/xfs_repair.c
>> +++ b/repair/xfs_repair.c
>> @@ -70,6 +70,7 @@ enum c_opt_nums {
>> CONVERT_FINOBT,
>> CONVERT_REFLINK,
>> CONVERT_RMAPBT,
>> + CONVERT_NREXT64,
>> C_MAX_OPTS,
>> };
>>
>> @@ -80,6 +81,7 @@ static char *c_opts[] = {
>> [CONVERT_FINOBT] = "finobt",
>> [CONVERT_REFLINK] = "reflink",
>> [CONVERT_RMAPBT] = "rmapbt",
>> + [CONVERT_NREXT64] = "nrext64",
>> [C_MAX_OPTS] = NULL,
>> };
>>
>> @@ -357,6 +359,15 @@ process_args(int argc, char **argv)
>> _("-c rmapbt only supports upgrades\n"));
>> add_rmapbt = true;
>> break;
>> + case CONVERT_NREXT64:
>> + if (!val)
>> + do_abort(
>> + _("-c nrext64 requires a parameter\n"));
>> + if (strtol(val, NULL, 0) != 1)
>> + do_abort(
>> + _("-c nrext64 only supports upgrades\n"));
>> + add_nrext64 = true;
>> + break;
>> default:
>> unknown('c', val);
>> break;
>> --
>> 2.30.2
>>
--
chandan
^ permalink raw reply [flat|nested] 34+ messages in thread* Re: [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature
2022-01-07 16:17 ` Chandan Babu R
@ 2022-01-07 19:03 ` Darrick J. Wong
2022-01-08 16:16 ` Chandan Babu R
0 siblings, 1 reply; 34+ messages in thread
From: Darrick J. Wong @ 2022-01-07 19:03 UTC (permalink / raw)
To: Chandan Babu R; +Cc: linux-xfs, david
On Fri, Jan 07, 2022 at 09:47:10PM +0530, Chandan Babu R wrote:
> On 05 Jan 2022 at 06:47, Darrick J. Wong wrote:
> > On Tue, Dec 14, 2021 at 02:18:10PM +0530, Chandan Babu R wrote:
> >> This commit adds support to xfs_repair to allow upgrading an existing
> >> filesystem to support per-inode large extent counters.
> >>
> >> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
> >> ---
> >> repair/globals.c | 1 +
> >> repair/globals.h | 1 +
> >> repair/phase2.c | 35 ++++++++++++++++++++++++++++++++++-
> >> repair/xfs_repair.c | 11 +++++++++++
> >> 4 files changed, 47 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/repair/globals.c b/repair/globals.c
> >> index d89507b1..2f29391a 100644
> >> --- a/repair/globals.c
> >> +++ b/repair/globals.c
> >> @@ -53,6 +53,7 @@ bool add_bigtime; /* add support for timestamps up to 2486 */
> >> bool add_finobt; /* add free inode btrees */
> >> bool add_reflink; /* add reference count btrees */
> >> bool add_rmapbt; /* add reverse mapping btrees */
> >> +bool add_nrext64;
> >>
> >> /* misc status variables */
> >>
> >> diff --git a/repair/globals.h b/repair/globals.h
> >> index 53ff2532..af0bcb6b 100644
> >> --- a/repair/globals.h
> >> +++ b/repair/globals.h
> >> @@ -94,6 +94,7 @@ extern bool add_bigtime; /* add support for timestamps up to 2486 */
> >> extern bool add_finobt; /* add free inode btrees */
> >> extern bool add_reflink; /* add reference count btrees */
> >> extern bool add_rmapbt; /* add reverse mapping btrees */
> >> +extern bool add_nrext64;
> >>
> >> /* misc status variables */
> >>
> >> diff --git a/repair/phase2.c b/repair/phase2.c
> >> index c811ed5d..c9db3281 100644
> >> --- a/repair/phase2.c
> >> +++ b/repair/phase2.c
> >> @@ -191,6 +191,7 @@ check_new_v5_geometry(
> >> struct xfs_perag *pag;
> >> xfs_agnumber_t agno;
> >> xfs_ino_t rootino;
> >> + uint old_bm_maxlevels[2];
> >> int min_logblocks;
> >> int error;
> >>
> >> @@ -201,6 +202,12 @@ check_new_v5_geometry(
> >> memcpy(&old_sb, &mp->m_sb, sizeof(struct xfs_sb));
> >> memcpy(&mp->m_sb, new_sb, sizeof(struct xfs_sb));
> >>
> >> + old_bm_maxlevels[0] = mp->m_bm_maxlevels[0];
> >> + old_bm_maxlevels[1] = mp->m_bm_maxlevels[1];
> >> +
> >> + xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
> >> + xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK);
> >
> > Ahh... I see why you added my (evil) patch that allows upgrading a
> > filesystem to reflink -- you need the check_new_v5_geometry function so
> > that you can check if the log size is big enough to handle larger bmbt
> > trees.
> >
> > Hmm, I guess I should work on separating this from the actual
> > rmap/reflink/finobt upgrade code, since I have no idea if we /ever/ want
> > to support that.
> >
>
> I can do that. I will include the trimmed down version of the patch before
> posting the patchset once again.
I separated that megapatch into smaller pieces yesterday, so I'll point
you to it once it all goes through QA.
--D
> > --D
> >
> >> +
> >> /* Do we have a big enough log? */
> >> min_logblocks = libxfs_log_calc_minimum_size(mp);
> >> if (old_sb.sb_logblocks < min_logblocks) {
> >> @@ -288,6 +295,9 @@ check_new_v5_geometry(
> >> pag->pagi_init = 0;
> >> }
> >>
> >> + mp->m_bm_maxlevels[0] = old_bm_maxlevels[0];
> >> + mp->m_bm_maxlevels[1] = old_bm_maxlevels[1];
> >> +
> >> /*
> >> * Put back the old superblock.
> >> */
> >> @@ -366,6 +376,28 @@ set_rmapbt(
> >> return true;
> >> }
> >>
> >> +static bool
> >> +set_nrext64(
> >> + struct xfs_mount *mp,
> >> + struct xfs_sb *new_sb)
> >> +{
> >> + if (!xfs_sb_version_hascrc(&mp->m_sb)) {
> >> + printf(
> >> + _("Nrext64 only supported on V5 filesystems.\n"));
> >> + exit(0);
> >> + }
> >> +
> >> + if (xfs_sb_version_hasnrext64(&mp->m_sb)) {
> >> + printf(_("Filesystem already supports nrext64.\n"));
> >> + exit(0);
> >> + }
> >> +
> >> + printf(_("Adding nrext64 to filesystem.\n"));
> >> + new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NREXT64;
> >> + new_sb->sb_features_incompat |= XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR;
> >> + return true;
> >> +}
> >> +
> >> /* Perform the user's requested upgrades on filesystem. */
> >> static void
> >> upgrade_filesystem(
> >> @@ -388,7 +420,8 @@ upgrade_filesystem(
> >> dirty |= set_reflink(mp, &new_sb);
> >> if (add_rmapbt)
> >> dirty |= set_rmapbt(mp, &new_sb);
> >> -
> >> + if (add_nrext64)
> >> + dirty |= set_nrext64(mp, &new_sb);
> >> if (!dirty)
> >> return;
> >>
> >> diff --git a/repair/xfs_repair.c b/repair/xfs_repair.c
> >> index e250a5bf..96c9bb56 100644
> >> --- a/repair/xfs_repair.c
> >> +++ b/repair/xfs_repair.c
> >> @@ -70,6 +70,7 @@ enum c_opt_nums {
> >> CONVERT_FINOBT,
> >> CONVERT_REFLINK,
> >> CONVERT_RMAPBT,
> >> + CONVERT_NREXT64,
> >> C_MAX_OPTS,
> >> };
> >>
> >> @@ -80,6 +81,7 @@ static char *c_opts[] = {
> >> [CONVERT_FINOBT] = "finobt",
> >> [CONVERT_REFLINK] = "reflink",
> >> [CONVERT_RMAPBT] = "rmapbt",
> >> + [CONVERT_NREXT64] = "nrext64",
> >> [C_MAX_OPTS] = NULL,
> >> };
> >>
> >> @@ -357,6 +359,15 @@ process_args(int argc, char **argv)
> >> _("-c rmapbt only supports upgrades\n"));
> >> add_rmapbt = true;
> >> break;
> >> + case CONVERT_NREXT64:
> >> + if (!val)
> >> + do_abort(
> >> + _("-c nrext64 requires a parameter\n"));
> >> + if (strtol(val, NULL, 0) != 1)
> >> + do_abort(
> >> + _("-c nrext64 only supports upgrades\n"));
> >> + add_nrext64 = true;
> >> + break;
> >> default:
> >> unknown('c', val);
> >> break;
> >> --
> >> 2.30.2
> >>
>
>
> --
> chandan
^ permalink raw reply [flat|nested] 34+ messages in thread* Re: [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature
2022-01-07 19:03 ` Darrick J. Wong
@ 2022-01-08 16:16 ` Chandan Babu R
2022-01-08 18:28 ` Darrick J. Wong
0 siblings, 1 reply; 34+ messages in thread
From: Chandan Babu R @ 2022-01-08 16:16 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: linux-xfs, david
On 08 Jan 2022 at 00:33, Darrick J. Wong wrote:
> On Fri, Jan 07, 2022 at 09:47:10PM +0530, Chandan Babu R wrote:
>> On 05 Jan 2022 at 06:47, Darrick J. Wong wrote:
>> > On Tue, Dec 14, 2021 at 02:18:10PM +0530, Chandan Babu R wrote:
>> >> This commit adds support to xfs_repair to allow upgrading an existing
>> >> filesystem to support per-inode large extent counters.
>> >>
>> >> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
>> >> ---
>> >> repair/globals.c | 1 +
>> >> repair/globals.h | 1 +
>> >> repair/phase2.c | 35 ++++++++++++++++++++++++++++++++++-
>> >> repair/xfs_repair.c | 11 +++++++++++
>> >> 4 files changed, 47 insertions(+), 1 deletion(-)
>> >>
>> >> diff --git a/repair/globals.c b/repair/globals.c
>> >> index d89507b1..2f29391a 100644
>> >> --- a/repair/globals.c
>> >> +++ b/repair/globals.c
>> >> @@ -53,6 +53,7 @@ bool add_bigtime; /* add support for timestamps up to 2486 */
>> >> bool add_finobt; /* add free inode btrees */
>> >> bool add_reflink; /* add reference count btrees */
>> >> bool add_rmapbt; /* add reverse mapping btrees */
>> >> +bool add_nrext64;
>> >>
>> >> /* misc status variables */
>> >>
>> >> diff --git a/repair/globals.h b/repair/globals.h
>> >> index 53ff2532..af0bcb6b 100644
>> >> --- a/repair/globals.h
>> >> +++ b/repair/globals.h
>> >> @@ -94,6 +94,7 @@ extern bool add_bigtime; /* add support for timestamps up to 2486 */
>> >> extern bool add_finobt; /* add free inode btrees */
>> >> extern bool add_reflink; /* add reference count btrees */
>> >> extern bool add_rmapbt; /* add reverse mapping btrees */
>> >> +extern bool add_nrext64;
>> >>
>> >> /* misc status variables */
>> >>
>> >> diff --git a/repair/phase2.c b/repair/phase2.c
>> >> index c811ed5d..c9db3281 100644
>> >> --- a/repair/phase2.c
>> >> +++ b/repair/phase2.c
>> >> @@ -191,6 +191,7 @@ check_new_v5_geometry(
>> >> struct xfs_perag *pag;
>> >> xfs_agnumber_t agno;
>> >> xfs_ino_t rootino;
>> >> + uint old_bm_maxlevels[2];
>> >> int min_logblocks;
>> >> int error;
>> >>
>> >> @@ -201,6 +202,12 @@ check_new_v5_geometry(
>> >> memcpy(&old_sb, &mp->m_sb, sizeof(struct xfs_sb));
>> >> memcpy(&mp->m_sb, new_sb, sizeof(struct xfs_sb));
>> >>
>> >> + old_bm_maxlevels[0] = mp->m_bm_maxlevels[0];
>> >> + old_bm_maxlevels[1] = mp->m_bm_maxlevels[1];
>> >> +
>> >> + xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
>> >> + xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK);
>> >
>> > Ahh... I see why you added my (evil) patch that allows upgrading a
>> > filesystem to reflink -- you need the check_new_v5_geometry function so
>> > that you can check if the log size is big enough to handle larger bmbt
>> > trees.
>> >
>> > Hmm, I guess I should work on separating this from the actual
>> > rmap/reflink/finobt upgrade code, since I have no idea if we /ever/ want
>> > to support that.
>> >
>>
>> I can do that. I will include the trimmed down version of the patch before
>> posting the patchset once again.
>
> I separated that megapatch into smaller pieces yesterday, so I'll point
> you to it once it all goes through QA.
>
Ok. I will wait.
--
chandan
^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature
2022-01-08 16:16 ` Chandan Babu R
@ 2022-01-08 18:28 ` Darrick J. Wong
2022-01-10 15:42 ` Chandan Babu R
0 siblings, 1 reply; 34+ messages in thread
From: Darrick J. Wong @ 2022-01-08 18:28 UTC (permalink / raw)
To: Chandan Babu R; +Cc: linux-xfs, david
On Sat, Jan 08, 2022 at 09:46:08PM +0530, Chandan Babu R wrote:
> On 08 Jan 2022 at 00:33, Darrick J. Wong wrote:
> > On Fri, Jan 07, 2022 at 09:47:10PM +0530, Chandan Babu R wrote:
> >> On 05 Jan 2022 at 06:47, Darrick J. Wong wrote:
> >> > On Tue, Dec 14, 2021 at 02:18:10PM +0530, Chandan Babu R wrote:
> >> >> This commit adds support to xfs_repair to allow upgrading an existing
> >> >> filesystem to support per-inode large extent counters.
> >> >>
> >> >> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
> >> >> ---
> >> >> repair/globals.c | 1 +
> >> >> repair/globals.h | 1 +
> >> >> repair/phase2.c | 35 ++++++++++++++++++++++++++++++++++-
> >> >> repair/xfs_repair.c | 11 +++++++++++
> >> >> 4 files changed, 47 insertions(+), 1 deletion(-)
> >> >>
> >> >> diff --git a/repair/globals.c b/repair/globals.c
> >> >> index d89507b1..2f29391a 100644
> >> >> --- a/repair/globals.c
> >> >> +++ b/repair/globals.c
> >> >> @@ -53,6 +53,7 @@ bool add_bigtime; /* add support for timestamps up to 2486 */
> >> >> bool add_finobt; /* add free inode btrees */
> >> >> bool add_reflink; /* add reference count btrees */
> >> >> bool add_rmapbt; /* add reverse mapping btrees */
> >> >> +bool add_nrext64;
> >> >>
> >> >> /* misc status variables */
> >> >>
> >> >> diff --git a/repair/globals.h b/repair/globals.h
> >> >> index 53ff2532..af0bcb6b 100644
> >> >> --- a/repair/globals.h
> >> >> +++ b/repair/globals.h
> >> >> @@ -94,6 +94,7 @@ extern bool add_bigtime; /* add support for timestamps up to 2486 */
> >> >> extern bool add_finobt; /* add free inode btrees */
> >> >> extern bool add_reflink; /* add reference count btrees */
> >> >> extern bool add_rmapbt; /* add reverse mapping btrees */
> >> >> +extern bool add_nrext64;
> >> >>
> >> >> /* misc status variables */
> >> >>
> >> >> diff --git a/repair/phase2.c b/repair/phase2.c
> >> >> index c811ed5d..c9db3281 100644
> >> >> --- a/repair/phase2.c
> >> >> +++ b/repair/phase2.c
> >> >> @@ -191,6 +191,7 @@ check_new_v5_geometry(
> >> >> struct xfs_perag *pag;
> >> >> xfs_agnumber_t agno;
> >> >> xfs_ino_t rootino;
> >> >> + uint old_bm_maxlevels[2];
> >> >> int min_logblocks;
> >> >> int error;
> >> >>
> >> >> @@ -201,6 +202,12 @@ check_new_v5_geometry(
> >> >> memcpy(&old_sb, &mp->m_sb, sizeof(struct xfs_sb));
> >> >> memcpy(&mp->m_sb, new_sb, sizeof(struct xfs_sb));
> >> >>
> >> >> + old_bm_maxlevels[0] = mp->m_bm_maxlevels[0];
> >> >> + old_bm_maxlevels[1] = mp->m_bm_maxlevels[1];
> >> >> +
> >> >> + xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
> >> >> + xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK);
> >> >
> >> > Ahh... I see why you added my (evil) patch that allows upgrading a
> >> > filesystem to reflink -- you need the check_new_v5_geometry function so
> >> > that you can check if the log size is big enough to handle larger bmbt
> >> > trees.
> >> >
> >> > Hmm, I guess I should work on separating this from the actual
> >> > rmap/reflink/finobt upgrade code, since I have no idea if we /ever/ want
> >> > to support that.
> >> >
> >>
> >> I can do that. I will include the trimmed down version of the patch before
> >> posting the patchset once again.
> >
> > I separated that megapatch into smaller pieces yesterday, so I'll point
> > you to it once it all goes through QA.
> >
>
> Ok. I will wait.
Here's one patch to fix a bug I found in the upgrade code, because
apparently we weren't resyncing the secondary superblocks
https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/commit/?h=upgrade-older-features&id=e0f4bff35adcae98943ee95701c207c628940d8f
And here's an updated version of xfs_repair infrastructure you need to
add nrext64, without the extraneous code to add other features. It also
now recomputes m_features and the maxlevels values so you don't have to
do that anymore.
https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/commit/?h=upgrade-older-features&id=acaf9c0355ee09da035845f15b4e44ba2ec24a6e
--D
>
> --
> chandan
^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature
2022-01-08 18:28 ` Darrick J. Wong
@ 2022-01-10 15:42 ` Chandan Babu R
0 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2022-01-10 15:42 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: linux-xfs, david
On 08 Jan 2022 at 23:58, Darrick J. Wong wrote:
> On Sat, Jan 08, 2022 at 09:46:08PM +0530, Chandan Babu R wrote:
>> On 08 Jan 2022 at 00:33, Darrick J. Wong wrote:
>> > On Fri, Jan 07, 2022 at 09:47:10PM +0530, Chandan Babu R wrote:
>> >> On 05 Jan 2022 at 06:47, Darrick J. Wong wrote:
>> >> > On Tue, Dec 14, 2021 at 02:18:10PM +0530, Chandan Babu R wrote:
>> >> >> This commit adds support to xfs_repair to allow upgrading an existing
>> >> >> filesystem to support per-inode large extent counters.
>> >> >>
>> >> >> Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
>> >> >> ---
>> >> >> repair/globals.c | 1 +
>> >> >> repair/globals.h | 1 +
>> >> >> repair/phase2.c | 35 ++++++++++++++++++++++++++++++++++-
>> >> >> repair/xfs_repair.c | 11 +++++++++++
>> >> >> 4 files changed, 47 insertions(+), 1 deletion(-)
>> >> >>
>> >> >> diff --git a/repair/globals.c b/repair/globals.c
>> >> >> index d89507b1..2f29391a 100644
>> >> >> --- a/repair/globals.c
>> >> >> +++ b/repair/globals.c
>> >> >> @@ -53,6 +53,7 @@ bool add_bigtime; /* add support for timestamps up to 2486 */
>> >> >> bool add_finobt; /* add free inode btrees */
>> >> >> bool add_reflink; /* add reference count btrees */
>> >> >> bool add_rmapbt; /* add reverse mapping btrees */
>> >> >> +bool add_nrext64;
>> >> >>
>> >> >> /* misc status variables */
>> >> >>
>> >> >> diff --git a/repair/globals.h b/repair/globals.h
>> >> >> index 53ff2532..af0bcb6b 100644
>> >> >> --- a/repair/globals.h
>> >> >> +++ b/repair/globals.h
>> >> >> @@ -94,6 +94,7 @@ extern bool add_bigtime; /* add support for timestamps up to 2486 */
>> >> >> extern bool add_finobt; /* add free inode btrees */
>> >> >> extern bool add_reflink; /* add reference count btrees */
>> >> >> extern bool add_rmapbt; /* add reverse mapping btrees */
>> >> >> +extern bool add_nrext64;
>> >> >>
>> >> >> /* misc status variables */
>> >> >>
>> >> >> diff --git a/repair/phase2.c b/repair/phase2.c
>> >> >> index c811ed5d..c9db3281 100644
>> >> >> --- a/repair/phase2.c
>> >> >> +++ b/repair/phase2.c
>> >> >> @@ -191,6 +191,7 @@ check_new_v5_geometry(
>> >> >> struct xfs_perag *pag;
>> >> >> xfs_agnumber_t agno;
>> >> >> xfs_ino_t rootino;
>> >> >> + uint old_bm_maxlevels[2];
>> >> >> int min_logblocks;
>> >> >> int error;
>> >> >>
>> >> >> @@ -201,6 +202,12 @@ check_new_v5_geometry(
>> >> >> memcpy(&old_sb, &mp->m_sb, sizeof(struct xfs_sb));
>> >> >> memcpy(&mp->m_sb, new_sb, sizeof(struct xfs_sb));
>> >> >>
>> >> >> + old_bm_maxlevels[0] = mp->m_bm_maxlevels[0];
>> >> >> + old_bm_maxlevels[1] = mp->m_bm_maxlevels[1];
>> >> >> +
>> >> >> + xfs_bmap_compute_maxlevels(mp, XFS_DATA_FORK);
>> >> >> + xfs_bmap_compute_maxlevels(mp, XFS_ATTR_FORK);
>> >> >
>> >> > Ahh... I see why you added my (evil) patch that allows upgrading a
>> >> > filesystem to reflink -- you need the check_new_v5_geometry function so
>> >> > that you can check if the log size is big enough to handle larger bmbt
>> >> > trees.
>> >> >
>> >> > Hmm, I guess I should work on separating this from the actual
>> >> > rmap/reflink/finobt upgrade code, since I have no idea if we /ever/ want
>> >> > to support that.
>> >> >
>> >>
>> >> I can do that. I will include the trimmed down version of the patch before
>> >> posting the patchset once again.
>> >
>> > I separated that megapatch into smaller pieces yesterday, so I'll point
>> > you to it once it all goes through QA.
>> >
>>
>> Ok. I will wait.
>
> Here's one patch to fix a bug I found in the upgrade code, because
> apparently we weren't resyncing the secondary superblocks
>
> https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/commit/?h=upgrade-older-features&id=e0f4bff35adcae98943ee95701c207c628940d8f
>
> And here's an updated version of xfs_repair infrastructure you need to
> add nrext64, without the extraneous code to add other features. It also
> now recomputes m_features and the maxlevels values so you don't have to
> do that anymore.
>
> https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfsprogs-dev.git/commit/?h=upgrade-older-features&id=acaf9c0355ee09da035845f15b4e44ba2ec24a6e
>
Thank you!
I have rebased my xfsprogs patchset on top of the above commit. I have also
removed the commits associated with atomic swapext feature though.
--
chandan
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH V4 20/20] xfsprogs: Define max extent length based on on-disk format definition
2021-12-14 8:47 [PATCH V4 00/20] xfsprogs: Extend per-inode extent counters Chandan Babu R
` (18 preceding siblings ...)
2021-12-14 8:48 ` [PATCH V4 19/20] xfsprogs: Add support for upgrading to NREXT64 feature Chandan Babu R
@ 2021-12-14 8:48 ` Chandan Babu R
19 siblings, 0 replies; 34+ messages in thread
From: Chandan Babu R @ 2021-12-14 8:48 UTC (permalink / raw)
To: linux-xfs; +Cc: Chandan Babu R, djwong, david
The maximum extent length depends on maximum block count that can be stored in
a BMBT record. Hence this commit defines MAXEXTLEN based on
BMBT_BLOCKCOUNT_BITLEN.
While at it, the commit also renames MAXEXTLEN to XFS_MAX_BMBT_EXTLEN.
Suggested-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Chandan Babu R <chandan.babu@oracle.com>
---
db/metadump.c | 2 +-
libxfs/xfs_bmap.c | 53 +++++++++++++++++++++--------------------
libxfs/xfs_format.h | 23 +++++++++---------
libxfs/xfs_inode_buf.c | 4 ++--
libxfs/xfs_trans_resv.c | 10 ++++----
mkfs/xfs_mkfs.c | 6 ++---
repair/phase4.c | 2 +-
7 files changed, 51 insertions(+), 49 deletions(-)
diff --git a/db/metadump.c b/db/metadump.c
index 8a9aec75..3ed4d81a 100644
--- a/db/metadump.c
+++ b/db/metadump.c
@@ -20,7 +20,7 @@
#include "field.h"
#include "dir2.h"
-#define DEFAULT_MAX_EXT_SIZE MAXEXTLEN
+#define DEFAULT_MAX_EXT_SIZE XFS_MAX_BMBT_EXTLEN
/* copy all metadata structures to/from a file */
diff --git a/libxfs/xfs_bmap.c b/libxfs/xfs_bmap.c
index 9dd24678..ac66a33d 100644
--- a/libxfs/xfs_bmap.c
+++ b/libxfs/xfs_bmap.c
@@ -536,7 +536,7 @@ __xfs_bmap_add_free(
ASSERT(bno != NULLFSBLOCK);
ASSERT(len > 0);
- ASSERT(len <= MAXEXTLEN);
+ ASSERT(len <= XFS_MAX_BMBT_EXTLEN);
ASSERT(!isnullstartblock(bno));
agno = XFS_FSB_TO_AGNO(mp, bno);
agbno = XFS_FSB_TO_AGBNO(mp, bno);
@@ -1493,7 +1493,7 @@ xfs_bmap_add_extent_delay_real(
LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
LEFT.br_state == new->br_state &&
- LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+ LEFT.br_blockcount + new->br_blockcount <= XFS_MAX_BMBT_EXTLEN)
state |= BMAP_LEFT_CONTIG;
/*
@@ -1511,13 +1511,13 @@ xfs_bmap_add_extent_delay_real(
new_endoff == RIGHT.br_startoff &&
new->br_startblock + new->br_blockcount == RIGHT.br_startblock &&
new->br_state == RIGHT.br_state &&
- new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
+ new->br_blockcount + RIGHT.br_blockcount <= XFS_MAX_BMBT_EXTLEN &&
((state & (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
BMAP_RIGHT_FILLING)) !=
(BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
BMAP_RIGHT_FILLING) ||
LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
- <= MAXEXTLEN))
+ <= XFS_MAX_BMBT_EXTLEN))
state |= BMAP_RIGHT_CONTIG;
error = 0;
@@ -2041,7 +2041,7 @@ xfs_bmap_add_extent_unwritten_real(
LEFT.br_startoff + LEFT.br_blockcount == new->br_startoff &&
LEFT.br_startblock + LEFT.br_blockcount == new->br_startblock &&
LEFT.br_state == new->br_state &&
- LEFT.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+ LEFT.br_blockcount + new->br_blockcount <= XFS_MAX_BMBT_EXTLEN)
state |= BMAP_LEFT_CONTIG;
/*
@@ -2059,13 +2059,13 @@ xfs_bmap_add_extent_unwritten_real(
new_endoff == RIGHT.br_startoff &&
new->br_startblock + new->br_blockcount == RIGHT.br_startblock &&
new->br_state == RIGHT.br_state &&
- new->br_blockcount + RIGHT.br_blockcount <= MAXEXTLEN &&
+ new->br_blockcount + RIGHT.br_blockcount <= XFS_MAX_BMBT_EXTLEN &&
((state & (BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
BMAP_RIGHT_FILLING)) !=
(BMAP_LEFT_CONTIG | BMAP_LEFT_FILLING |
BMAP_RIGHT_FILLING) ||
LEFT.br_blockcount + new->br_blockcount + RIGHT.br_blockcount
- <= MAXEXTLEN))
+ <= XFS_MAX_BMBT_EXTLEN))
state |= BMAP_RIGHT_CONTIG;
/*
@@ -2551,15 +2551,15 @@ xfs_bmap_add_extent_hole_delay(
*/
if ((state & BMAP_LEFT_VALID) && (state & BMAP_LEFT_DELAY) &&
left.br_startoff + left.br_blockcount == new->br_startoff &&
- left.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+ left.br_blockcount + new->br_blockcount <= XFS_MAX_BMBT_EXTLEN)
state |= BMAP_LEFT_CONTIG;
if ((state & BMAP_RIGHT_VALID) && (state & BMAP_RIGHT_DELAY) &&
new->br_startoff + new->br_blockcount == right.br_startoff &&
- new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
+ new->br_blockcount + right.br_blockcount <= XFS_MAX_BMBT_EXTLEN &&
(!(state & BMAP_LEFT_CONTIG) ||
(left.br_blockcount + new->br_blockcount +
- right.br_blockcount <= MAXEXTLEN)))
+ right.br_blockcount <= XFS_MAX_BMBT_EXTLEN)))
state |= BMAP_RIGHT_CONTIG;
/*
@@ -2702,17 +2702,17 @@ xfs_bmap_add_extent_hole_real(
left.br_startoff + left.br_blockcount == new->br_startoff &&
left.br_startblock + left.br_blockcount == new->br_startblock &&
left.br_state == new->br_state &&
- left.br_blockcount + new->br_blockcount <= MAXEXTLEN)
+ left.br_blockcount + new->br_blockcount <= XFS_MAX_BMBT_EXTLEN)
state |= BMAP_LEFT_CONTIG;
if ((state & BMAP_RIGHT_VALID) && !(state & BMAP_RIGHT_DELAY) &&
new->br_startoff + new->br_blockcount == right.br_startoff &&
new->br_startblock + new->br_blockcount == right.br_startblock &&
new->br_state == right.br_state &&
- new->br_blockcount + right.br_blockcount <= MAXEXTLEN &&
+ new->br_blockcount + right.br_blockcount <= XFS_MAX_BMBT_EXTLEN &&
(!(state & BMAP_LEFT_CONTIG) ||
left.br_blockcount + new->br_blockcount +
- right.br_blockcount <= MAXEXTLEN))
+ right.br_blockcount <= XFS_MAX_BMBT_EXTLEN))
state |= BMAP_RIGHT_CONTIG;
error = 0;
@@ -2947,15 +2947,15 @@ xfs_bmap_extsize_align(
/*
* For large extent hint sizes, the aligned extent might be larger than
- * MAXEXTLEN. In that case, reduce the size by an extsz so that it pulls
- * the length back under MAXEXTLEN. The outer allocation loops handle
+ * XFS_MAX_BMBT_EXTLEN. In that case, reduce the size by an extsz so that it pulls
+ * the length back under XFS_MAX_BMBT_EXTLEN. The outer allocation loops handle
* short allocation just fine, so it is safe to do this. We only want to
* do it when we are forced to, though, because it means more allocation
* operations are required.
*/
- while (align_alen > MAXEXTLEN)
+ while (align_alen > XFS_MAX_BMBT_EXTLEN)
align_alen -= extsz;
- ASSERT(align_alen <= MAXEXTLEN);
+ ASSERT(align_alen <= XFS_MAX_BMBT_EXTLEN);
/*
* If the previous block overlaps with this proposed allocation
@@ -3045,9 +3045,9 @@ xfs_bmap_extsize_align(
return -EINVAL;
} else {
ASSERT(orig_off >= align_off);
- /* see MAXEXTLEN handling above */
+ /* see XFS_MAX_BMBT_EXTLEN handling above */
ASSERT(orig_end <= align_off + align_alen ||
- align_alen + extsz > MAXEXTLEN);
+ align_alen + extsz > XFS_MAX_BMBT_EXTLEN);
}
#ifdef DEBUG
@@ -4012,7 +4012,7 @@ xfs_bmapi_reserve_delalloc(
* Cap the alloc length. Keep track of prealloc so we know whether to
* tag the inode before we return.
*/
- alen = XFS_FILBLKS_MIN(len + prealloc, MAXEXTLEN);
+ alen = XFS_FILBLKS_MIN(len + prealloc, XFS_MAX_BMBT_EXTLEN);
if (!eof)
alen = XFS_FILBLKS_MIN(alen, got->br_startoff - aoff);
if (prealloc && alen >= len)
@@ -4145,7 +4145,7 @@ xfs_bmapi_allocate(
if (!xfs_iext_peek_prev_extent(ifp, &bma->icur, &bma->prev))
bma->prev.br_startoff = NULLFILEOFF;
} else {
- bma->length = XFS_FILBLKS_MIN(bma->length, MAXEXTLEN);
+ bma->length = XFS_FILBLKS_MIN(bma->length, XFS_MAX_BMBT_EXTLEN);
if (!bma->eof)
bma->length = XFS_FILBLKS_MIN(bma->length,
bma->got.br_startoff - bma->offset);
@@ -4465,8 +4465,8 @@ xfs_bmapi_write(
* xfs_extlen_t and therefore 32 bits. Hence we have to
* check for 32-bit overflows and handle them here.
*/
- if (len > (xfs_filblks_t)MAXEXTLEN)
- bma.length = MAXEXTLEN;
+ if (len > (xfs_filblks_t)XFS_MAX_BMBT_EXTLEN)
+ bma.length = XFS_MAX_BMBT_EXTLEN;
else
bma.length = len;
@@ -4601,7 +4601,8 @@ xfs_bmapi_convert_delalloc(
bma.ip = ip;
bma.wasdel = true;
bma.offset = bma.got.br_startoff;
- bma.length = max_t(xfs_filblks_t, bma.got.br_blockcount, MAXEXTLEN);
+ bma.length = max_t(xfs_filblks_t, bma.got.br_blockcount,
+ XFS_MAX_BMBT_EXTLEN);
bma.minleft = xfs_bmapi_minleft(tp, ip, whichfork);
/*
@@ -4682,7 +4683,7 @@ xfs_bmapi_remap(
ifp = XFS_IFORK_PTR(ip, whichfork);
ASSERT(len > 0);
- ASSERT(len <= (xfs_filblks_t)MAXEXTLEN);
+ ASSERT(len <= (xfs_filblks_t)XFS_MAX_BMBT_EXTLEN);
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
ASSERT(!(flags & ~(XFS_BMAPI_ATTRFORK | XFS_BMAPI_PREALLOC |
XFS_BMAPI_NORMAP)));
@@ -5682,7 +5683,7 @@ xfs_bmse_can_merge(
if ((left->br_startoff + left->br_blockcount != startoff) ||
(left->br_startblock + left->br_blockcount != got->br_startblock) ||
(left->br_state != got->br_state) ||
- (left->br_blockcount + got->br_blockcount > MAXEXTLEN))
+ (left->br_blockcount + got->br_blockcount > XFS_MAX_BMBT_EXTLEN))
return false;
return true;
diff --git a/libxfs/xfs_format.h b/libxfs/xfs_format.h
index 0ed9d5ac..ac6aa23e 100644
--- a/libxfs/xfs_format.h
+++ b/libxfs/xfs_format.h
@@ -1073,15 +1073,6 @@ enum xfs_dinode_fmt {
{ XFS_DINODE_FMT_BTREE, "btree" }, \
{ XFS_DINODE_FMT_UUID, "uuid" }
-/*
- * Max values for extlen, extnum, aextnum.
- */
-#define MAXEXTLEN ((xfs_extlen_t)0x1fffff) /* 21 bits */
-#define XFS_MAX_EXTCNT_DATA_FORK ((xfs_extnum_t)0xffffffffffff) /* Unsigned 48-bits */
-#define XFS_MAX_EXTCNT_ATTR_FORK ((xfs_aextnum_t)0xffffffff) /* Unsigned 32-bits */
-#define XFS_MAX_EXTCNT_DATA_FORK_OLD ((xfs_extnum_t)0x7fffffff) /* Signed 32-bits */
-#define XFS_MAX_EXTCNT_ATTR_FORK_OLD ((xfs_aextnum_t)0x7fff) /* Signed 16-bits */
-
/*
* Inode minimum and maximum sizes.
*/
@@ -1810,8 +1801,18 @@ typedef struct xfs_bmdr_block {
#define BMBT_STARTBLOCK_BITLEN 52
#define BMBT_BLOCKCOUNT_BITLEN 21
-#define BMBT_STARTOFF_MASK ((1ULL << BMBT_STARTOFF_BITLEN) - 1)
-#define BMBT_BLOCKCOUNT_MASK ((1ULL << BMBT_BLOCKCOUNT_BITLEN) - 1)
+#define BMBT_STARTOFF_MASK ((1ULL << BMBT_STARTOFF_BITLEN) - 1)
+
+/*
+ * Max values for extlen, extnum, aextnum.
+ */
+#define XFS_MAX_BMBT_EXTLEN ((xfs_extlen_t)(1ULL << BMBT_BLOCKCOUNT_BITLEN) - 1)
+#define XFS_MAX_EXTCNT_DATA_FORK ((xfs_extnum_t)0xffffffffffff) /* Unsigned 48-bits */
+#define XFS_MAX_EXTCNT_ATTR_FORK ((xfs_aextnum_t)0xffffffff) /* Unsigned 32-bits */
+#define XFS_MAX_EXTCNT_DATA_FORK_OLD ((xfs_extnum_t)0x7fffffff) /* Signed 32-bits */
+#define XFS_MAX_EXTCNT_ATTR_FORK_OLD ((xfs_aextnum_t)0x7fff) /* Signed 16-bits */
+
+#define BMBT_BLOCKCOUNT_MASK XFS_MAX_BMBT_EXTLEN
/*
* bmbt records have a file offset (block) field that is 54 bits wide, so this
diff --git a/libxfs/xfs_inode_buf.c b/libxfs/xfs_inode_buf.c
index a484e14b..601ca604 100644
--- a/libxfs/xfs_inode_buf.c
+++ b/libxfs/xfs_inode_buf.c
@@ -666,7 +666,7 @@ xfs_inode_validate_extsize(
if (extsize_bytes % blocksize_bytes)
return __this_address;
- if (extsize > MAXEXTLEN)
+ if (extsize > XFS_MAX_BMBT_EXTLEN)
return __this_address;
if (!rt_flag && extsize > mp->m_sb.sb_agblocks / 2)
@@ -723,7 +723,7 @@ xfs_inode_validate_cowextsize(
if (cowextsize_bytes % mp->m_sb.sb_blocksize)
return __this_address;
- if (cowextsize > MAXEXTLEN)
+ if (cowextsize > XFS_MAX_BMBT_EXTLEN)
return __this_address;
if (cowextsize > mp->m_sb.sb_agblocks / 2)
diff --git a/libxfs/xfs_trans_resv.c b/libxfs/xfs_trans_resv.c
index 4fd2c62b..559879e5 100644
--- a/libxfs/xfs_trans_resv.c
+++ b/libxfs/xfs_trans_resv.c
@@ -198,8 +198,8 @@ xfs_calc_inode_chunk_res(
/*
* Per-extent log reservation for the btree changes involved in freeing or
* allocating a realtime extent. We have to be able to log as many rtbitmap
- * blocks as needed to mark inuse MAXEXTLEN blocks' worth of realtime extents,
- * as well as the realtime summary block.
+ * blocks as needed to mark inuse XFS_MAX_BMBT_EXTLEN blocks' worth of realtime
+ * extents, as well as the realtime summary block.
*/
static unsigned int
xfs_rtalloc_log_count(
@@ -209,7 +209,7 @@ xfs_rtalloc_log_count(
unsigned int blksz = XFS_FSB_TO_B(mp, 1);
unsigned int rtbmp_bytes;
- rtbmp_bytes = (MAXEXTLEN / mp->m_sb.sb_rextsize) / NBBY;
+ rtbmp_bytes = (XFS_MAX_BMBT_EXTLEN / mp->m_sb.sb_rextsize) / NBBY;
return (howmany(rtbmp_bytes, blksz) + 1) * num_ops;
}
@@ -246,7 +246,7 @@ xfs_rtalloc_log_count(
* the inode's bmap btree: max depth * block size
* the agfs of the ags from which the extents are allocated: 2 * sector
* the superblock free block counter: sector size
- * the realtime bitmap: ((MAXEXTLEN / rtextsize) / NBBY) bytes
+ * the realtime bitmap: ((XFS_MAX_BMBT_EXTLEN / rtextsize) / NBBY) bytes
* the realtime summary: 1 block
* the allocation btrees: 2 trees * (2 * max depth - 1) * block size
* And the bmap_finish transaction can free bmap blocks in a join (t3):
@@ -298,7 +298,7 @@ xfs_calc_write_reservation(
* the agf for each of the ags: 2 * sector size
* the agfl for each of the ags: 2 * sector size
* the super block to reflect the freed blocks: sector size
- * the realtime bitmap: 2 exts * ((MAXEXTLEN / rtextsize) / NBBY) bytes
+ * the realtime bitmap: 2 exts * ((XFS_MAX_BMBT_EXTLEN / rtextsize) / NBBY) bytes
* the realtime summary: 2 exts * 1 block
* worst case split in allocation btrees per extent assuming 2 extents:
* 2 exts * 2 trees * (2 * max depth - 1) * block size
diff --git a/mkfs/xfs_mkfs.c b/mkfs/xfs_mkfs.c
index 6609776f..deb3b070 100644
--- a/mkfs/xfs_mkfs.c
+++ b/mkfs/xfs_mkfs.c
@@ -2399,7 +2399,7 @@ validate_extsize_hint(
fprintf(stderr,
_("illegal extent size hint %lld, must be less than %u.\n"),
(long long)cli->fsx.fsx_extsize,
- min(MAXEXTLEN, mp->m_sb.sb_agblocks / 2));
+ min(XFS_MAX_BMBT_EXTLEN, mp->m_sb.sb_agblocks / 2));
usage();
}
@@ -2422,7 +2422,7 @@ _("illegal extent size hint %lld, must be less than %u.\n"),
fprintf(stderr,
_("illegal extent size hint %lld, must be less than %u and a multiple of %u.\n"),
(long long)cli->fsx.fsx_extsize,
- min(MAXEXTLEN, mp->m_sb.sb_agblocks / 2),
+ min(XFS_MAX_BMBT_EXTLEN, mp->m_sb.sb_agblocks / 2),
mp->m_sb.sb_rextsize);
usage();
}
@@ -2451,7 +2451,7 @@ validate_cowextsize_hint(
fprintf(stderr,
_("illegal CoW extent size hint %lld, must be less than %u.\n"),
(long long)cli->fsx.fsx_cowextsize,
- min(MAXEXTLEN, mp->m_sb.sb_agblocks / 2));
+ min(XFS_MAX_BMBT_EXTLEN, mp->m_sb.sb_agblocks / 2));
usage();
}
}
diff --git a/repair/phase4.c b/repair/phase4.c
index eb043002..8cad2b80 100644
--- a/repair/phase4.c
+++ b/repair/phase4.c
@@ -372,7 +372,7 @@ phase4(xfs_mount_t *mp)
if (rt_start == 0) {
rt_start = bno;
rt_len = 1;
- } else if (rt_len == MAXEXTLEN) {
+ } else if (rt_len == XFS_MAX_BMBT_EXTLEN) {
/*
* large extent case
*/
--
2.30.2
^ permalink raw reply related [flat|nested] 34+ messages in thread