ceph: update support for PGID64, PGPOOL3, OSDENC protocol features
Support (and require) the PGID64, PGPOOL3, and OSDENC protocol features.
These have been present in ceph.git since v0.42, Feb 2012. Require these
features to simplify support; nobody is running older userspace.
Note that the new request and reply encoding is still not in place, so the new
code is not yet functional.
Signed-off-by: Sage Weil <sage@inktank.com>
Reviewed-by: Alex Elder <elder@inktank.com>
diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c
index 81118db..9119193 100644
--- a/net/ceph/osdmap.c
+++ b/net/ceph/osdmap.c
@@ -45,13 +45,8 @@
*/
static void calc_pg_masks(struct ceph_pg_pool_info *pi)
{
- pi->pg_num_mask = (1 << calc_bits_of(le32_to_cpu(pi->v.pg_num)-1)) - 1;
- pi->pgp_num_mask =
- (1 << calc_bits_of(le32_to_cpu(pi->v.pgp_num)-1)) - 1;
- pi->lpg_num_mask =
- (1 << calc_bits_of(le32_to_cpu(pi->v.lpg_num)-1)) - 1;
- pi->lpgp_num_mask =
- (1 << calc_bits_of(le32_to_cpu(pi->v.lpgp_num)-1)) - 1;
+ pi->pg_num_mask = (1 << calc_bits_of(pi->pg_num-1)) - 1;
+ pi->pgp_num_mask = (1 << calc_bits_of(pi->pgp_num-1)) - 1;
}
/*
@@ -452,7 +447,7 @@
return 0;
}
-static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id)
+static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, u64 id)
{
struct ceph_pg_pool_info *pi;
struct rb_node *n = root->rb_node;
@@ -508,24 +503,57 @@
static int __decode_pool(void **p, void *end, struct ceph_pg_pool_info *pi)
{
- unsigned int n, m;
+ u8 ev, cv;
+ unsigned len, num;
+ void *pool_end;
- ceph_decode_copy(p, &pi->v, sizeof(pi->v));
- calc_pg_masks(pi);
+ ceph_decode_need(p, end, 2 + 4, bad);
+ ev = ceph_decode_8(p); /* encoding version */
+ cv = ceph_decode_8(p); /* compat version */
+ if (ev < 5) {
+ pr_warning("got v %d < 5 cv %d of ceph_pg_pool\n", ev, cv);
+ return -EINVAL;
+ }
+ if (cv > 7) {
+ pr_warning("got v %d cv %d > 7 of ceph_pg_pool\n", ev, cv);
+ return -EINVAL;
+ }
+ len = ceph_decode_32(p);
+ ceph_decode_need(p, end, len, bad);
+ pool_end = *p + len;
- /* num_snaps * snap_info_t */
- n = le32_to_cpu(pi->v.num_snaps);
- while (n--) {
- ceph_decode_need(p, end, sizeof(u64) + 1 + sizeof(u64) +
- sizeof(struct ceph_timespec), bad);
- *p += sizeof(u64) + /* key */
- 1 + sizeof(u64) + /* u8, snapid */
- sizeof(struct ceph_timespec);
- m = ceph_decode_32(p); /* snap name */
- *p += m;
+ pi->type = ceph_decode_8(p);
+ pi->size = ceph_decode_8(p);
+ pi->crush_ruleset = ceph_decode_8(p);
+ pi->object_hash = ceph_decode_8(p);
+
+ pi->pg_num = ceph_decode_32(p);
+ pi->pgp_num = ceph_decode_32(p);
+
+ *p += 4 + 4; /* skip lpg* */
+ *p += 4; /* skip last_change */
+ *p += 8 + 4; /* skip snap_seq, snap_epoch */
+
+ /* skip snaps */
+ num = ceph_decode_32(p);
+ while (num--) {
+ *p += 8; /* snapid key */
+ *p += 1 + 1; /* versions */
+ len = ceph_decode_32(p);
+ *p += len;
}
- *p += le32_to_cpu(pi->v.num_removed_snap_intervals) * sizeof(u64) * 2;
+ /* skip removed snaps */
+ num = ceph_decode_32(p);
+ *p += num * (8 + 8);
+
+ *p += 8; /* skip auid */
+ pi->flags = ceph_decode_64(p);
+
+ /* ignore the rest */
+
+ *p = pool_end;
+ calc_pg_masks(pi);
return 0;
bad:
@@ -535,14 +563,15 @@
static int __decode_pool_names(void **p, void *end, struct ceph_osdmap *map)
{
struct ceph_pg_pool_info *pi;
- u32 num, len, pool;
+ u32 num, len;
+ u64 pool;
ceph_decode_32_safe(p, end, num, bad);
dout(" %d pool names\n", num);
while (num--) {
- ceph_decode_32_safe(p, end, pool, bad);
+ ceph_decode_64_safe(p, end, pool, bad);
ceph_decode_32_safe(p, end, len, bad);
- dout(" pool %d len %d\n", pool, len);
+ dout(" pool %llu len %d\n", pool, len);
ceph_decode_need(p, end, len, bad);
pi = __lookup_pg_pool(&map->pg_pools, pool);
if (pi) {
@@ -633,7 +662,6 @@
struct ceph_osdmap *map;
u16 version;
u32 len, max, i;
- u8 ev;
int err = -EINVAL;
void *start = *p;
struct ceph_pg_pool_info *pi;
@@ -646,9 +674,12 @@
map->pg_temp = RB_ROOT;
ceph_decode_16_safe(p, end, version, bad);
- if (version > CEPH_OSDMAP_VERSION) {
- pr_warning("got unknown v %d > %d of osdmap\n", version,
- CEPH_OSDMAP_VERSION);
+ if (version > 6) {
+ pr_warning("got unknown v %d > 6 of osdmap\n", version);
+ goto bad;
+ }
+ if (version < 6) {
+ pr_warning("got old v %d < 6 of osdmap\n", version);
goto bad;
}
@@ -660,20 +691,12 @@
ceph_decode_32_safe(p, end, max, bad);
while (max--) {
- ceph_decode_need(p, end, 4 + 1 + sizeof(pi->v), bad);
+ ceph_decode_need(p, end, 8 + 2, bad);
err = -ENOMEM;
pi = kzalloc(sizeof(*pi), GFP_NOFS);
if (!pi)
goto bad;
- pi->id = ceph_decode_32(p);
- err = -EINVAL;
- ev = ceph_decode_8(p); /* encoding version */
- if (ev > CEPH_PG_POOL_VERSION) {
- pr_warning("got unknown v %d > %d of ceph_pg_pool\n",
- ev, CEPH_PG_POOL_VERSION);
- kfree(pi);
- goto bad;
- }
+ pi->id = ceph_decode_64(p);
err = __decode_pool(p, end, pi);
if (err < 0) {
kfree(pi);
@@ -682,12 +705,10 @@
__insert_pg_pool(&map->pg_pools, pi);
}
- if (version >= 5) {
- err = __decode_pool_names(p, end, map);
- if (err < 0) {
- dout("fail to decode pool names");
- goto bad;
- }
+ err = __decode_pool_names(p, end, map);
+ if (err < 0) {
+ dout("fail to decode pool names");
+ goto bad;
}
ceph_decode_32_safe(p, end, map->pool_max, bad);
@@ -788,16 +809,17 @@
struct ceph_fsid fsid;
u32 epoch = 0;
struct ceph_timespec modified;
- u32 len, pool;
- __s32 new_pool_max, new_flags, max;
+ s32 len;
+ u64 pool;
+ __s64 new_pool_max;
+ __s32 new_flags, max;
void *start = *p;
int err = -EINVAL;
u16 version;
ceph_decode_16_safe(p, end, version, bad);
- if (version > CEPH_OSDMAP_INC_VERSION) {
- pr_warning("got unknown v %d > %d of inc osdmap\n", version,
- CEPH_OSDMAP_INC_VERSION);
+ if (version > 6) {
+ pr_warning("got unknown v %d > %d of inc osdmap\n", version, 6);
goto bad;
}
@@ -807,7 +829,7 @@
epoch = ceph_decode_32(p);
BUG_ON(epoch != map->epoch+1);
ceph_decode_copy(p, &modified, sizeof(modified));
- new_pool_max = ceph_decode_32(p);
+ new_pool_max = ceph_decode_64(p);
new_flags = ceph_decode_32(p);
/* full map? */
@@ -857,18 +879,9 @@
/* new_pool */
ceph_decode_32_safe(p, end, len, bad);
while (len--) {
- __u8 ev;
struct ceph_pg_pool_info *pi;
- ceph_decode_32_safe(p, end, pool, bad);
- ceph_decode_need(p, end, 1 + sizeof(pi->v), bad);
- ev = ceph_decode_8(p); /* encoding version */
- if (ev > CEPH_PG_POOL_VERSION) {
- pr_warning("got unknown v %d > %d of ceph_pg_pool\n",
- ev, CEPH_PG_POOL_VERSION);
- err = -EINVAL;
- goto bad;
- }
+ ceph_decode_64_safe(p, end, pool, bad);
pi = __lookup_pg_pool(&map->pg_pools, pool);
if (!pi) {
pi = kzalloc(sizeof(*pi), GFP_NOFS);
@@ -894,7 +907,7 @@
while (len--) {
struct ceph_pg_pool_info *pi;
- ceph_decode_32_safe(p, end, pool, bad);
+ ceph_decode_64_safe(p, end, pool, bad);
pi = __lookup_pg_pool(&map->pg_pools, pool);
if (pi)
__remove_pg_pool(&map->pg_pools, pi);
@@ -1097,8 +1110,8 @@
pool = __lookup_pg_pool(&osdmap->pg_pools, pgid.pool);
if (!pool)
return -EIO;
- pgid.seed = ceph_str_hash(pool->v.object_hash, oid, strlen(oid));
- num = le32_to_cpu(pool->v.pg_num);
+ pgid.seed = ceph_str_hash(pool->object_hash, oid, strlen(oid));
+ num = pool->pg_num;
num_mask = pool->pg_num_mask;
dout("calc_object_layout '%s' pgid %lld.%x\n", oid, pgid.pool,
@@ -1132,8 +1145,7 @@
return NULL;
/* pg_temp? */
- t = ceph_stable_mod(ps, le32_to_cpu(pool->v.pg_num),
- pool->pgp_num_mask);
+ t = ceph_stable_mod(ps, pool->pg_num, pool->pgp_num_mask);
pgid.seed = t;
pg = __lookup_pg_mapping(&osdmap->pg_temp, pgid);
if (pg) {
@@ -1142,26 +1154,24 @@
}
/* crush */
- ruleno = crush_find_rule(osdmap->crush, pool->v.crush_ruleset,
- pool->v.type, pool->v.size);
+ ruleno = crush_find_rule(osdmap->crush, pool->crush_ruleset,
+ pool->type, pool->size);
if (ruleno < 0) {
pr_err("no crush rule pool %d ruleset %d type %d size %d\n",
- poolid, pool->v.crush_ruleset, pool->v.type,
- pool->v.size);
+ poolid, pool->crush_ruleset, pool->type,
+ pool->size);
return NULL;
}
- pps = ceph_stable_mod(ps,
- le32_to_cpu(pool->v.pgp_num),
- pool->pgp_num_mask);
+ pps = ceph_stable_mod(ps, pool->pgp_num, pool->pgp_num_mask);
pps += poolid;
r = crush_do_rule(osdmap->crush, ruleno, pps, osds,
- min_t(int, pool->v.size, *num),
+ min_t(int, pool->size, *num),
osdmap->osd_weight);
if (r < 0) {
pr_err("error %d from crush rule: pool %d ruleset %d type %d"
- " size %d\n", r, poolid, pool->v.crush_ruleset,
- pool->v.type, pool->v.size);
+ " size %d\n", r, poolid, pool->crush_ruleset,
+ pool->type, pool->size);
return NULL;
}
*num = r;