/**
* Compute file level page index by stripe level page offset
*/
pgoff_t lov_stripe_pgoff(struct lov_stripe_md *lsm, pgoff_t stripe_index,
int stripe)
{
obd_off offset;
offset = lov_stripe_size(lsm, stripe_index << PAGE_CACHE_SHIFT,
stripe);
return offset >> PAGE_CACHE_SHIFT;
}
void lov_merge_attrs(struct obdo *tgt, struct obdo *src, u64 valid,
struct lov_stripe_md *lsm, int stripeno, int *set)
{
valid &= src->o_valid;
if (*set) {
if (valid & OBD_MD_FLSIZE) {
/* this handles sparse files properly */
u64 lov_size;
lov_size = lov_stripe_size(lsm, src->o_size, stripeno);
if (lov_size > tgt->o_size)
tgt->o_size = lov_size;
}
if (valid & OBD_MD_FLBLOCKS)
tgt->o_blocks += src->o_blocks;
if (valid & OBD_MD_FLBLKSZ)
tgt->o_blksize += src->o_blksize;
if (valid & OBD_MD_FLCTIME && tgt->o_ctime < src->o_ctime)
tgt->o_ctime = src->o_ctime;
if (valid & OBD_MD_FLMTIME && tgt->o_mtime < src->o_mtime)
tgt->o_mtime = src->o_mtime;
if (valid & OBD_MD_FLDATAVERSION)
tgt->o_data_version += src->o_data_version;
} else {
memcpy(tgt, src, sizeof(*tgt));
tgt->o_oi = lsm->lsm_oi;
if (valid & OBD_MD_FLSIZE)
tgt->o_size = lov_stripe_size(lsm, src->o_size,
stripeno);
}
/* data_version needs to be valid on all stripes to be correct! */
if (!(valid & OBD_MD_FLDATAVERSION))
tgt->o_valid &= ~OBD_MD_FLDATAVERSION;
*set += 1;
}
/** Merge the lock value block(&lvb) attributes and KMS from each of the
* stripes in a file into a single lvb. It is expected that the caller
* initializes the current atime, mtime, ctime to avoid regressing a more
* uptodate time on the local client.
*/
int lov_merge_lvb_kms(struct lov_stripe_md *lsm,
struct ost_lvb *lvb, __u64 *kms_place)
{
__u64 size = 0;
__u64 kms = 0;
__u64 blocks = 0;
s64 current_mtime = lvb->lvb_mtime;
s64 current_atime = lvb->lvb_atime;
s64 current_ctime = lvb->lvb_ctime;
int i;
int rc = 0;
assert_spin_locked(&lsm->lsm_lock);
LASSERT(lsm->lsm_lock_owner == current_pid());
CDEBUG(D_INODE, "MDT ID "DOSTID" initial value: s=%llu m=%llu a=%llu c=%llu b=%llu\n",
POSTID(&lsm->lsm_oi), lvb->lvb_size, lvb->lvb_mtime,
lvb->lvb_atime, lvb->lvb_ctime, lvb->lvb_blocks);
for (i = 0; i < lsm->lsm_stripe_count; i++) {
struct lov_oinfo *loi = lsm->lsm_oinfo[i];
u64 lov_size, tmpsize;
if (OST_LVB_IS_ERR(loi->loi_lvb.lvb_blocks)) {
rc = OST_LVB_GET_ERR(loi->loi_lvb.lvb_blocks);
continue;
}
tmpsize = loi->loi_kms;
lov_size = lov_stripe_size(lsm, tmpsize, i);
if (lov_size > kms)
kms = lov_size;
if (loi->loi_lvb.lvb_size > tmpsize)
tmpsize = loi->loi_lvb.lvb_size;
lov_size = lov_stripe_size(lsm, tmpsize, i);
if (lov_size > size)
size = lov_size;
/* merge blocks, mtime, atime */
blocks += loi->loi_lvb.lvb_blocks;
if (loi->loi_lvb.lvb_mtime > current_mtime)
current_mtime = loi->loi_lvb.lvb_mtime;
if (loi->loi_lvb.lvb_atime > current_atime)
current_atime = loi->loi_lvb.lvb_atime;
if (loi->loi_lvb.lvb_ctime > current_ctime)
current_ctime = loi->loi_lvb.lvb_ctime;
CDEBUG(D_INODE, "MDT ID "DOSTID" on OST[%u]: s=%llu m=%llu a=%llu c=%llu b=%llu\n",
POSTID(&lsm->lsm_oi), loi->loi_ost_idx,
loi->loi_lvb.lvb_size, loi->loi_lvb.lvb_mtime,
loi->loi_lvb.lvb_atime, loi->loi_lvb.lvb_ctime,
loi->loi_lvb.lvb_blocks);
}
*kms_place = kms;
lvb->lvb_size = size;
lvb->lvb_blocks = blocks;
lvb->lvb_mtime = current_mtime;
lvb->lvb_atime = current_atime;
lvb->lvb_ctime = current_ctime;
return rc;
}
/**
* Break down the FIEMAP request and send appropriate calls to individual OSTs.
* This also handles the restarting of FIEMAP calls in case mapping overflows
* the available number of extents in single call.
*
* \param env [in] lustre environment
* \param obj [in] file object
* \param fmkey [in] fiemap request header and other info
* \param fiemap [out] fiemap buffer holding retrived map extents
* \param buflen [in/out] max buffer length of @fiemap, when iterate
* each OST, it is used to limit max map needed
* \retval 0 success
* \retval < 0 error
*/
static int lov_object_fiemap(const struct lu_env *env, struct cl_object *obj,
struct ll_fiemap_info_key *fmkey,
struct fiemap *fiemap, size_t *buflen)
{
struct lov_stripe_md *lsm;
struct cl_object *subobj = NULL;
struct lov_obd *lov = lu2lov_dev(obj->co_lu.lo_dev)->ld_lov;
struct fiemap *fm_local = NULL;
struct fiemap_extent *lcl_fm_ext;
loff_t fm_start;
loff_t fm_end;
loff_t fm_length;
loff_t fm_end_offset;
int count_local;
int ost_index = 0;
int start_stripe;
int current_extent = 0;
int rc = 0;
int last_stripe;
int cur_stripe = 0;
int cur_stripe_wrap = 0;
int stripe_count;
unsigned int buffer_size = FIEMAP_BUFFER_SIZE;
/* Whether have we collected enough extents */
bool enough = false;
/* EOF for object */
bool ost_eof = false;
/* done with required mapping for this OST? */
bool ost_done = false;
ENTRY;
lsm = lov_lsm_addref(cl2lov(obj));
if (lsm == NULL)
RETURN(-ENODATA);
/**
* If the stripe_count > 1 and the application does not understand
* DEVICE_ORDER flag, it cannot interpret the extents correctly.
*/
if (lsm->lsm_stripe_count > 1 && !(fiemap->fm_flags &
FIEMAP_FLAG_DEVICE_ORDER))
GOTO(out_lsm, rc = -ENOTSUPP);
if (lsm_is_released(lsm)) {
if (fiemap->fm_start < fmkey->lfik_oa.o_size) {
/**
* released file, return a minimal FIEMAP if
* request fits in file-size.
*/
fiemap->fm_mapped_extents = 1;
fiemap->fm_extents[0].fe_logical = fiemap->fm_start;
if (fiemap->fm_start + fiemap->fm_length <
fmkey->lfik_oa.o_size)
fiemap->fm_extents[0].fe_length =
fiemap->fm_length;
else
fiemap->fm_extents[0].fe_length =
fmkey->lfik_oa.o_size -
fiemap->fm_start;
fiemap->fm_extents[0].fe_flags |=
FIEMAP_EXTENT_UNKNOWN | FIEMAP_EXTENT_LAST;
}
GOTO(out_lsm, rc = 0);
}
if (fiemap_count_to_size(fiemap->fm_extent_count) < buffer_size)
buffer_size = fiemap_count_to_size(fiemap->fm_extent_count);
OBD_ALLOC_LARGE(fm_local, buffer_size);
if (fm_local == NULL)
GOTO(out_lsm, rc = -ENOMEM);
lcl_fm_ext = &fm_local->fm_extents[0];
count_local = fiemap_size_to_count(buffer_size);
fm_start = fiemap->fm_start;
fm_length = fiemap->fm_length;
/* Calculate start stripe, last stripe and length of mapping */
start_stripe = lov_stripe_number(lsm, fm_start);
fm_end = (fm_length == ~0ULL) ? fmkey->lfik_oa.o_size :
fm_start + fm_length - 1;
/* If fm_length != ~0ULL but fm_start_fm_length-1 exceeds file size */
if (fm_end > fmkey->lfik_oa.o_size)
fm_end = fmkey->lfik_oa.o_size;
last_stripe = fiemap_calc_last_stripe(lsm, fm_start, fm_end,
start_stripe, &stripe_count);
fm_end_offset = fiemap_calc_fm_end_offset(fiemap, lsm, fm_start, fm_end,
&start_stripe);
if (fm_end_offset == -EINVAL)
GOTO(out_fm_local, rc = -EINVAL);
/**
* Requested extent count exceeds the fiemap buffer size, shrink our
* ambition.
*/
if (fiemap_count_to_size(fiemap->fm_extent_count) > *buflen)
fiemap->fm_extent_count = fiemap_size_to_count(*buflen);
if (fiemap->fm_extent_count == 0)
count_local = 0;
/* Check each stripe */
for (cur_stripe = start_stripe; stripe_count > 0;
--stripe_count,
cur_stripe = (cur_stripe + 1) % lsm->lsm_stripe_count) {
loff_t req_fm_len; /* Stores length of required mapping */
loff_t len_mapped_single_call;
loff_t lun_start;
loff_t lun_end;
loff_t obd_object_end;
unsigned int ext_count;
cur_stripe_wrap = cur_stripe;
/* Find out range of mapping on this stripe */
if ((lov_stripe_intersects(lsm, cur_stripe, fm_start, fm_end,
&lun_start, &obd_object_end)) == 0)
continue;
if (lov_oinfo_is_dummy(lsm->lsm_oinfo[cur_stripe]))
GOTO(out_fm_local, rc = -EIO);
/* If this is a continuation FIEMAP call and we are on
* starting stripe then lun_start needs to be set to
* fm_end_offset */
if (fm_end_offset != 0 && cur_stripe == start_stripe)
lun_start = fm_end_offset;
if (fm_length != ~0ULL) {
/* Handle fm_start + fm_length overflow */
if (fm_start + fm_length < fm_start)
fm_length = ~0ULL - fm_start;
lun_end = lov_size_to_stripe(lsm, fm_start + fm_length,
cur_stripe);
} else {
lun_end = ~0ULL;
}
if (lun_start == lun_end)
continue;
req_fm_len = obd_object_end - lun_start;
fm_local->fm_length = 0;
len_mapped_single_call = 0;
/* find lobsub object */
subobj = lov_find_subobj(env, cl2lov(obj), lsm,
cur_stripe);
if (IS_ERR(subobj))
GOTO(out_fm_local, rc = PTR_ERR(subobj));
/* If the output buffer is very large and the objects have many
* extents we may need to loop on a single OST repeatedly */
ost_eof = false;
ost_done = false;
do {
if (fiemap->fm_extent_count > 0) {
/* Don't get too many extents. */
if (current_extent + count_local >
fiemap->fm_extent_count)
count_local = fiemap->fm_extent_count -
current_extent;
}
lun_start += len_mapped_single_call;
fm_local->fm_length = req_fm_len -
len_mapped_single_call;
req_fm_len = fm_local->fm_length;
fm_local->fm_extent_count = enough ? 1 : count_local;
fm_local->fm_mapped_extents = 0;
fm_local->fm_flags = fiemap->fm_flags;
ost_index = lsm->lsm_oinfo[cur_stripe]->loi_ost_idx;
if (ost_index < 0 ||
ost_index >= lov->desc.ld_tgt_count)
GOTO(obj_put, rc = -EINVAL);
/* If OST is inactive, return extent with UNKNOWN
* flag. */
if (!lov->lov_tgts[ost_index]->ltd_active) {
fm_local->fm_flags |= FIEMAP_EXTENT_LAST;
fm_local->fm_mapped_extents = 1;
lcl_fm_ext[0].fe_logical = lun_start;
lcl_fm_ext[0].fe_length = obd_object_end -
lun_start;
lcl_fm_ext[0].fe_flags |= FIEMAP_EXTENT_UNKNOWN;
goto inactive_tgt;
}
fm_local->fm_start = lun_start;
fm_local->fm_flags &= ~FIEMAP_FLAG_DEVICE_ORDER;
memcpy(&fmkey->lfik_fiemap, fm_local,
sizeof(*fm_local));
*buflen = fiemap_count_to_size(
fm_local->fm_extent_count);
rc = cl_object_fiemap(env, subobj, fmkey, fm_local,
buflen);
if (rc != 0)
GOTO(obj_put, rc);
inactive_tgt:
ext_count = fm_local->fm_mapped_extents;
if (ext_count == 0) {
ost_done = true;
/* If last stripe has hold at the end,
* we need to return */
if (cur_stripe_wrap == last_stripe) {
fiemap->fm_mapped_extents = 0;
goto finish;
}
break;
} else if (enough) {
/*
* We've collected enough extents and there are
* more extents after it.
*/
goto finish;
}
/* If we just need num of extents, got to next device */
if (fiemap->fm_extent_count == 0) {
current_extent += ext_count;
break;
}
/* prepare to copy retrived map extents */
len_mapped_single_call =
lcl_fm_ext[ext_count - 1].fe_logical -
lun_start + lcl_fm_ext[ext_count - 1].fe_length;
/* Have we finished mapping on this device? */
if (req_fm_len <= len_mapped_single_call)
ost_done = true;
/* Clear the EXTENT_LAST flag which can be present on
* the last extent */
if (lcl_fm_ext[ext_count - 1].fe_flags &
FIEMAP_EXTENT_LAST)
lcl_fm_ext[ext_count - 1].fe_flags &=
~FIEMAP_EXTENT_LAST;
if (lov_stripe_size(lsm,
lcl_fm_ext[ext_count - 1].fe_logical +
lcl_fm_ext[ext_count - 1].fe_length,
cur_stripe) >= fmkey->lfik_oa.o_size)
ost_eof = true;
fiemap_prepare_and_copy_exts(fiemap, lcl_fm_ext,
ost_index, ext_count,
current_extent);
current_extent += ext_count;
/* Ran out of available extents? */
if (current_extent >= fiemap->fm_extent_count)
enough = true;
} while (!ost_done && !ost_eof);
cl_object_put(env, subobj);
subobj = NULL;
if (cur_stripe_wrap == last_stripe)
goto finish;
} /* for each stripe */
finish:
/* Indicate that we are returning device offsets unless file just has
* single stripe */
if (lsm->lsm_stripe_count > 1)
fiemap->fm_flags |= FIEMAP_FLAG_DEVICE_ORDER;
if (fiemap->fm_extent_count == 0)
goto skip_last_device_calc;
/* Check if we have reached the last stripe and whether mapping for that
* stripe is done. */
if ((cur_stripe_wrap == last_stripe) && (ost_done || ost_eof))
fiemap->fm_extents[current_extent - 1].fe_flags |=
FIEMAP_EXTENT_LAST;
skip_last_device_calc:
fiemap->fm_mapped_extents = current_extent;
obj_put:
if (subobj != NULL)
cl_object_put(env, subobj);
out_fm_local:
OBD_FREE_LARGE(fm_local, buffer_size);
out_lsm:
lov_lsm_put(lsm);
return rc;
}
/** Merge the lock value block(&lvb) attributes and KMS from each of the
* stripes in a file into a single lvb. It is expected that the caller
* initializes the current atime, mtime, ctime to avoid regressing a more
* uptodate time on the local client.
*/
int lov_merge_lvb_kms(struct lov_stripe_md *lsm,
struct ost_lvb *lvb, __u64 *kms_place)
{
__u64 size = 0;
__u64 kms = 0;
__u64 blocks = 0;
obd_time current_mtime = lvb->lvb_mtime;
obd_time current_atime = lvb->lvb_atime;
obd_time current_ctime = lvb->lvb_ctime;
int i;
int rc = 0;
LASSERT_SPIN_LOCKED(&lsm->lsm_lock);
#ifdef __KERNEL__
LASSERT(lsm->lsm_lock_owner == cfs_curproc_pid());
#endif
CDEBUG(D_INODE, "MDT FID "DFID" initial value: s="LPU64" m="LPU64
" a="LPU64" c="LPU64" b="LPU64"\n",
lsm->lsm_object_seq, (__u32)lsm->lsm_object_id,
(__u32)(lsm->lsm_object_id >> 32), lvb->lvb_size,
lvb->lvb_mtime, lvb->lvb_atime, lvb->lvb_ctime, lvb->lvb_blocks);
for (i = 0; i < lsm->lsm_stripe_count; i++) {
struct lov_oinfo *loi = lsm->lsm_oinfo[i];
obd_size lov_size, tmpsize;
if (OST_LVB_IS_ERR(loi->loi_lvb.lvb_blocks)) {
rc = OST_LVB_GET_ERR(loi->loi_lvb.lvb_blocks);
continue;
}
tmpsize = loi->loi_kms;
lov_size = lov_stripe_size(lsm, tmpsize, i);
if (lov_size > kms)
kms = lov_size;
if (loi->loi_lvb.lvb_size > tmpsize)
tmpsize = loi->loi_lvb.lvb_size;
lov_size = lov_stripe_size(lsm, tmpsize, i);
if (lov_size > size)
size = lov_size;
/* merge blocks, mtime, atime */
blocks += loi->loi_lvb.lvb_blocks;
if (loi->loi_lvb.lvb_mtime > current_mtime)
current_mtime = loi->loi_lvb.lvb_mtime;
if (loi->loi_lvb.lvb_atime > current_atime)
current_atime = loi->loi_lvb.lvb_atime;
if (loi->loi_lvb.lvb_ctime > current_ctime)
current_ctime = loi->loi_lvb.lvb_ctime;
CDEBUG(D_INODE, "MDT FID "DFID" on OST[%u]: s="LPU64" m="LPU64
" a="LPU64" c="LPU64" b="LPU64"\n",
lsm->lsm_object_seq, (__u32)lsm->lsm_object_id,
(__u32)(lsm->lsm_object_id >> 32), loi->loi_ost_idx,
loi->loi_lvb.lvb_size, loi->loi_lvb.lvb_mtime,
loi->loi_lvb.lvb_atime, loi->loi_lvb.lvb_ctime,
loi->loi_lvb.lvb_blocks);
}
*kms_place = kms;
lvb->lvb_size = size;
lvb->lvb_blocks = blocks;
lvb->lvb_mtime = current_mtime;
lvb->lvb_atime = current_atime;
lvb->lvb_ctime = current_ctime;
RETURN(rc);
}
