summaryrefslogtreecommitdiffstats
path: root/net/ceph/mon_client.c
diff options
context:
space:
mode:
authorIlya Dryomov <idryomov@gmail.com>2020-10-30 13:30:51 +0100
committerIlya Dryomov <idryomov@gmail.com>2020-12-14 23:21:50 +0100
commita5cbd5fc22d5043a8a76e15d75d031fe24d1f69c (patch)
tree2379298419dcbfaddef2d7bbd1cba86ba68c8bd4 /net/ceph/mon_client.c
parentlibceph: factor out finish_auth() (diff)
downloadlinux-a5cbd5fc22d5043a8a76e15d75d031fe24d1f69c.tar.xz
linux-a5cbd5fc22d5043a8a76e15d75d031fe24d1f69c.zip
libceph, ceph: get and handle cluster maps with addrvecs
In preparation for msgr2, make the cluster send us maps with addrvecs including both LEGACY and MSGR2 addrs instead of a single LEGACY addr. This means advertising support for SERVER_NAUTILUS and also some older features: SERVER_MIMIC, MONENC and MONNAMES. MONNAMES and MONENC are actually pre-argonaut, we just never updated ceph_monmap_decode() for them. Decoding is unconditional, see commit 23c625ce3065 ("libceph: assume argonaut on the server side"). SERVER_MIMIC doesn't bear any meaning for the kernel client. Since ceph_decode_entity_addrvec() is guarded by encoding version checks (and in msgr2 case it is guarded implicitly by the fact that server is speaking msgr2), we assume MSG_ADDR2 for it. Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Diffstat (limited to 'net/ceph/mon_client.c')
-rw-r--r--net/ceph/mon_client.c145
1 files changed, 105 insertions, 40 deletions
diff --git a/net/ceph/mon_client.c b/net/ceph/mon_client.c
index ebfecf8d0918..a9754a7fa78c 100644
--- a/net/ceph/mon_client.c
+++ b/net/ceph/mon_client.c
@@ -36,57 +36,122 @@ static const struct ceph_connection_operations mon_con_ops;
static int __validate_auth(struct ceph_mon_client *monc);
+static int decode_mon_info(void **p, void *end, bool msgr2,
+ struct ceph_entity_addr *addr)
+{
+ void *mon_info_end;
+ u32 struct_len;
+ u8 struct_v;
+ int ret;
+
+ ret = ceph_start_decoding(p, end, 1, "mon_info_t", &struct_v,
+ &struct_len);
+ if (ret)
+ return ret;
+
+ mon_info_end = *p + struct_len;
+ ceph_decode_skip_string(p, end, e_inval); /* skip mon name */
+ ret = ceph_decode_entity_addrvec(p, end, msgr2, addr);
+ if (ret)
+ return ret;
+
+ *p = mon_info_end;
+ return 0;
+
+e_inval:
+ return -EINVAL;
+}
+
/*
* Decode a monmap blob (e.g., during mount).
+ *
+ * Assume MonMap v3 (i.e. encoding with MONNAMES and MONENC).
*/
-static struct ceph_monmap *ceph_monmap_decode(void *p, void *end)
+static struct ceph_monmap *ceph_monmap_decode(void **p, void *end, bool msgr2)
{
- struct ceph_monmap *m = NULL;
- int i, err = -EINVAL;
+ struct ceph_monmap *monmap = NULL;
struct ceph_fsid fsid;
- u32 epoch, num_mon;
- u32 len;
+ u32 struct_len;
+ int blob_len;
+ int num_mon;
+ u8 struct_v;
+ u32 epoch;
+ int ret;
+ int i;
+
+ ceph_decode_32_safe(p, end, blob_len, e_inval);
+ ceph_decode_need(p, end, blob_len, e_inval);
+
+ ret = ceph_start_decoding(p, end, 6, "monmap", &struct_v, &struct_len);
+ if (ret)
+ goto fail;
+
+ dout("%s struct_v %d\n", __func__, struct_v);
+ ceph_decode_copy_safe(p, end, &fsid, sizeof(fsid), e_inval);
+ ceph_decode_32_safe(p, end, epoch, e_inval);
+ if (struct_v >= 6) {
+ u32 feat_struct_len;
+ u8 feat_struct_v;
- ceph_decode_32_safe(&p, end, len, bad);
- ceph_decode_need(&p, end, len, bad);
+ *p += sizeof(struct ceph_timespec); /* skip last_changed */
+ *p += sizeof(struct ceph_timespec); /* skip created */
- dout("monmap_decode %p %p len %d (%d)\n", p, end, len, (int)(end-p));
- p += sizeof(u16); /* skip version */
+ ret = ceph_start_decoding(p, end, 1, "mon_feature_t",
+ &feat_struct_v, &feat_struct_len);
+ if (ret)
+ goto fail;
- ceph_decode_need(&p, end, sizeof(fsid) + 2*sizeof(u32), bad);
- ceph_decode_copy(&p, &fsid, sizeof(fsid));
- epoch = ceph_decode_32(&p);
+ *p += feat_struct_len; /* skip persistent_features */
- num_mon = ceph_decode_32(&p);
+ ret = ceph_start_decoding(p, end, 1, "mon_feature_t",
+ &feat_struct_v, &feat_struct_len);
+ if (ret)
+ goto fail;
+ *p += feat_struct_len; /* skip optional_features */
+ }
+ ceph_decode_32_safe(p, end, num_mon, e_inval);
+
+ dout("%s fsid %pU epoch %u num_mon %d\n", __func__, &fsid, epoch,
+ num_mon);
if (num_mon > CEPH_MAX_MON)
- goto bad;
- m = kmalloc(struct_size(m, mon_inst, num_mon), GFP_NOFS);
- if (m == NULL)
- return ERR_PTR(-ENOMEM);
- m->fsid = fsid;
- m->epoch = epoch;
- m->num_mon = num_mon;
- for (i = 0; i < num_mon; ++i) {
- struct ceph_entity_inst *inst = &m->mon_inst[i];
-
- /* copy name portion */
- ceph_decode_copy_safe(&p, end, &inst->name,
- sizeof(inst->name), bad);
- err = ceph_decode_entity_addr(&p, end, &inst->addr);
- if (err)
- goto bad;
+ goto e_inval;
+
+ monmap = kmalloc(struct_size(monmap, mon_inst, num_mon), GFP_NOIO);
+ if (!monmap) {
+ ret = -ENOMEM;
+ goto fail;
}
- dout("monmap_decode epoch %d, num_mon %d\n", m->epoch,
- m->num_mon);
- for (i = 0; i < m->num_mon; i++)
- dout("monmap_decode mon%d is %s\n", i,
- ceph_pr_addr(&m->mon_inst[i].addr));
- return m;
-bad:
- dout("monmap_decode failed with %d\n", err);
- kfree(m);
- return ERR_PTR(err);
+ monmap->fsid = fsid;
+ monmap->epoch = epoch;
+ monmap->num_mon = num_mon;
+
+ /* legacy_mon_addr map or mon_info map */
+ for (i = 0; i < num_mon; i++) {
+ struct ceph_entity_inst *inst = &monmap->mon_inst[i];
+
+ ceph_decode_skip_string(p, end, e_inval); /* skip mon name */
+ inst->name.type = CEPH_ENTITY_TYPE_MON;
+ inst->name.num = cpu_to_le64(i);
+
+ if (struct_v >= 6)
+ ret = decode_mon_info(p, end, msgr2, &inst->addr);
+ else
+ ret = ceph_decode_entity_addr(p, end, &inst->addr);
+ if (ret)
+ goto fail;
+
+ dout("%s mon%d addr %s\n", __func__, i,
+ ceph_pr_addr(&inst->addr));
+ }
+
+ return monmap;
+
+e_inval:
+ ret = -EINVAL;
+fail:
+ kfree(monmap);
+ return ERR_PTR(ret);
}
/*
@@ -476,7 +541,7 @@ static void ceph_monc_handle_map(struct ceph_mon_client *monc,
p = msg->front.iov_base;
end = p + msg->front.iov_len;
- monmap = ceph_monmap_decode(p, end);
+ monmap = ceph_monmap_decode(&p, end, false);
if (IS_ERR(monmap)) {
pr_err("problem decoding monmap, %d\n",
(int)PTR_ERR(monmap));