#include <linux/fs.h>
#include <linux/file.h>
#include <linux/slab.h>
-#include <linux/connector.h>
#include <linux/blkpg.h>
#include <linux/cpumask.h>
#include "drbd_int.h"
#include "drbd_req.h"
#include "drbd_wrappers.h"
#include <asm/unaligned.h>
-#include <linux/drbd_tag_magic.h>
#include <linux/drbd_limits.h>
-#include <linux/compiler.h>
#include <linux/kthread.h>
-static unsigned short *tl_add_blob(unsigned short *, enum drbd_tags, const void *, int);
-static unsigned short *tl_add_str(unsigned short *, enum drbd_tags, const char *);
-static unsigned short *tl_add_int(unsigned short *, enum drbd_tags, const void *);
-
-/* see get_sb_bdev and bd_claim */
+#include <net/genetlink.h>
+
+/* .doit */
+// int drbd_adm_create_resource(struct sk_buff *skb, struct genl_info *info);
+// int drbd_adm_delete_resource(struct sk_buff *skb, struct genl_info *info);
+
+int drbd_adm_add_minor(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_delete_minor(struct sk_buff *skb, struct genl_info *info);
+
+int drbd_adm_create_connection(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_delete_connection(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_down(struct sk_buff *skb, struct genl_info *info);
+
+int drbd_adm_set_role(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_attach(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_detach(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_connect(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_net_opts(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_resize(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_start_ov(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_new_c_uuid(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_disconnect(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_invalidate(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_invalidate_peer(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_pause_sync(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_resume_sync(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_suspend_io(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_resume_io(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_outdate(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_resource_opts(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_get_status(struct sk_buff *skb, struct genl_info *info);
+int drbd_adm_get_timeout_type(struct sk_buff *skb, struct genl_info *info);
+/* .dumpit */
+int drbd_adm_get_status_all(struct sk_buff *skb, struct netlink_callback *cb);
+
+#include <linux/drbd_genl_api.h>
+#include <linux/genl_magic_func.h>
+
+/* used blkdev_get_by_path, to claim our meta data device(s) */
static char *drbd_m_holder = "Hands off! this is DRBD's meta data device.";
-/* Generate the tag_list to struct functions */
-#define NL_PACKET(name, number, fields) \
-static int name ## _from_tags(struct drbd_conf *mdev, \
- unsigned short *tags, struct name *arg) __attribute__ ((unused)); \
-static int name ## _from_tags(struct drbd_conf *mdev, \
- unsigned short *tags, struct name *arg) \
-{ \
- int tag; \
- int dlen; \
- \
- while ((tag = get_unaligned(tags++)) != TT_END) { \
- dlen = get_unaligned(tags++); \
- switch (tag_number(tag)) { \
- fields \
- default: \
- if (tag & T_MANDATORY) { \
- dev_err(DEV, "Unknown tag: %d\n", tag_number(tag)); \
- return 0; \
- } \
- } \
- tags = (unsigned short *)((char *)tags + dlen); \
- } \
- return 1; \
-}
-#define NL_INTEGER(pn, pr, member) \
- case pn: /* D_ASSERT( tag_type(tag) == TT_INTEGER ); */ \
- arg->member = get_unaligned((int *)(tags)); \
- break;
-#define NL_INT64(pn, pr, member) \
- case pn: /* D_ASSERT( tag_type(tag) == TT_INT64 ); */ \
- arg->member = get_unaligned((u64 *)(tags)); \
- break;
-#define NL_BIT(pn, pr, member) \
- case pn: /* D_ASSERT( tag_type(tag) == TT_BIT ); */ \
- arg->member = *(char *)(tags) ? 1 : 0; \
- break;
-#define NL_STRING(pn, pr, member, len) \
- case pn: /* D_ASSERT( tag_type(tag) == TT_STRING ); */ \
- if (dlen > len) { \
- dev_err(DEV, "arg too long: %s (%u wanted, max len: %u bytes)\n", \
- #member, dlen, (unsigned int)len); \
- return 0; \
- } \
- arg->member ## _len = dlen; \
- memcpy(arg->member, tags, min_t(size_t, dlen, len)); \
- break;
-#include "linux/drbd_nl.h"
-
-/* Generate the struct to tag_list functions */
-#define NL_PACKET(name, number, fields) \
-static unsigned short* \
-name ## _to_tags(struct drbd_conf *mdev, \
- struct name *arg, unsigned short *tags) __attribute__ ((unused)); \
-static unsigned short* \
-name ## _to_tags(struct drbd_conf *mdev, \
- struct name *arg, unsigned short *tags) \
-{ \
- fields \
- return tags; \
-}
-
-#define NL_INTEGER(pn, pr, member) \
- put_unaligned(pn | pr | TT_INTEGER, tags++); \
- put_unaligned(sizeof(int), tags++); \
- put_unaligned(arg->member, (int *)tags); \
- tags = (unsigned short *)((char *)tags+sizeof(int));
-#define NL_INT64(pn, pr, member) \
- put_unaligned(pn | pr | TT_INT64, tags++); \
- put_unaligned(sizeof(u64), tags++); \
- put_unaligned(arg->member, (u64 *)tags); \
- tags = (unsigned short *)((char *)tags+sizeof(u64));
-#define NL_BIT(pn, pr, member) \
- put_unaligned(pn | pr | TT_BIT, tags++); \
- put_unaligned(sizeof(char), tags++); \
- *(char *)tags = arg->member; \
- tags = (unsigned short *)((char *)tags+sizeof(char));
-#define NL_STRING(pn, pr, member, len) \
- put_unaligned(pn | pr | TT_STRING, tags++); \
- put_unaligned(arg->member ## _len, tags++); \
- memcpy(tags, arg->member, arg->member ## _len); \
- tags = (unsigned short *)((char *)tags + arg->member ## _len);
-#include "linux/drbd_nl.h"
-
-void drbd_bcast_ev_helper(struct drbd_conf *mdev, char *helper_name);
-void drbd_nl_send_reply(struct cn_msg *, int);
+/* Configuration is strictly serialized, because generic netlink message
+ * processing is strictly serialized by the genl_lock().
+ * Which means we can use one static global drbd_config_context struct.
+ */
+static struct drbd_config_context {
+ /* assigned from drbd_genlmsghdr */
+ unsigned int minor;
+ /* assigned from request attributes, if present */
+ unsigned int volume;
+#define VOLUME_UNSPECIFIED (-1U)
+ /* pointer into the request skb,
+ * limited lifetime! */
+ char *conn_name;
+
+ /* reply buffer */
+ struct sk_buff *reply_skb;
+ /* pointer into reply buffer */
+ struct drbd_genlmsghdr *reply_dh;
+ /* resolved from attributes, if possible */
+ struct drbd_conf *mdev;
+ struct drbd_tconn *tconn;
+} adm_ctx;
-int drbd_khelper(struct drbd_conf *mdev, char *cmd)
+static void drbd_adm_send_reply(struct sk_buff *skb, struct genl_info *info)
{
- char *envp[] = { "HOME=/",
- "TERM=linux",
- "PATH=/sbin:/usr/sbin:/bin:/usr/bin",
- NULL, /* Will be set to address family */
- NULL, /* Will be set to address */
- NULL };
+ genlmsg_end(skb, genlmsg_data(nlmsg_data(nlmsg_hdr(skb))));
+ if (genlmsg_reply(skb, info))
+ printk(KERN_ERR "drbd: error sending genl reply\n");
+}
- char mb[12], af[20], ad[60], *afs;
- char *argv[] = {usermode_helper, cmd, mb, NULL };
- int ret;
+/* Used on a fresh "drbd_adm_prepare"d reply_skb, this cannot fail: The only
+ * reason it could fail was no space in skb, and there are 4k available. */
+int drbd_msg_put_info(const char *info)
+{
+ struct sk_buff *skb = adm_ctx.reply_skb;
+ struct nlattr *nla;
+ int err = -EMSGSIZE;
- snprintf(mb, 12, "minor-%d", mdev_to_minor(mdev));
+ if (!info || !info[0])
+ return 0;
+
+ nla = nla_nest_start(skb, DRBD_NLA_CFG_REPLY);
+ if (!nla)
+ return err;
+
+ err = nla_put_string(skb, T_info_text, info);
+ if (err) {
+ nla_nest_cancel(skb, nla);
+ return err;
+ } else
+ nla_nest_end(skb, nla);
+ return 0;
+}
+
+/* This would be a good candidate for a "pre_doit" hook,
+ * and per-family private info->pointers.
+ * But we need to stay compatible with older kernels.
+ * If it returns successfully, adm_ctx members are valid.
+ */
+#define DRBD_ADM_NEED_MINOR 1
+#define DRBD_ADM_NEED_CONN 2
+static int drbd_adm_prepare(struct sk_buff *skb, struct genl_info *info,
+ unsigned flags)
+{
+ struct drbd_genlmsghdr *d_in = info->userhdr;
+ const u8 cmd = info->genlhdr->cmd;
+ int err;
+
+ memset(&adm_ctx, 0, sizeof(adm_ctx));
+
+ /* genl_rcv_msg only checks for CAP_NET_ADMIN on "GENL_ADMIN_PERM" :( */
+ if (cmd != DRBD_ADM_GET_STATUS
+ && security_netlink_recv(skb, CAP_SYS_ADMIN))
+ return -EPERM;
+
+ adm_ctx.reply_skb = genlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
+ if (!adm_ctx.reply_skb)
+ goto fail;
+
+ adm_ctx.reply_dh = genlmsg_put_reply(adm_ctx.reply_skb,
+ info, &drbd_genl_family, 0, cmd);
+ /* put of a few bytes into a fresh skb of >= 4k will always succeed.
+ * but anyways */
+ if (!adm_ctx.reply_dh)
+ goto fail;
+
+ adm_ctx.reply_dh->minor = d_in->minor;
+ adm_ctx.reply_dh->ret_code = NO_ERROR;
+
+ if (info->attrs[DRBD_NLA_CFG_CONTEXT]) {
+ struct nlattr *nla;
+ /* parse and validate only */
+ err = drbd_cfg_context_from_attrs(NULL, info);
+ if (err)
+ goto fail;
+
+ /* It was present, and valid,
+ * copy it over to the reply skb. */
+ err = nla_put_nohdr(adm_ctx.reply_skb,
+ info->attrs[DRBD_NLA_CFG_CONTEXT]->nla_len,
+ info->attrs[DRBD_NLA_CFG_CONTEXT]);
+ if (err)
+ goto fail;
+
+ /* and assign stuff to the global adm_ctx */
+ nla = nested_attr_tb[__nla_type(T_ctx_volume)];
+ adm_ctx.volume = nla ? nla_get_u32(nla) : VOLUME_UNSPECIFIED;
+ nla = nested_attr_tb[__nla_type(T_ctx_conn_name)];
+ if (nla)
+ adm_ctx.conn_name = nla_data(nla);
+ } else
+ adm_ctx.volume = VOLUME_UNSPECIFIED;
+
+ adm_ctx.minor = d_in->minor;
+ adm_ctx.mdev = minor_to_mdev(d_in->minor);
+ adm_ctx.tconn = conn_by_name(adm_ctx.conn_name);
+
+ if (!adm_ctx.mdev && (flags & DRBD_ADM_NEED_MINOR)) {
+ drbd_msg_put_info("unknown minor");
+ return ERR_MINOR_INVALID;
+ }
+ if (!adm_ctx.tconn && (flags & DRBD_ADM_NEED_CONN)) {
+ drbd_msg_put_info("unknown connection");
+ return ERR_INVALID_REQUEST;
+ }
+
+ /* some more paranoia, if the request was over-determined */
+ if (adm_ctx.mdev && adm_ctx.tconn &&
+ adm_ctx.mdev->tconn != adm_ctx.tconn) {
+ pr_warning("request: minor=%u, conn=%s; but that minor belongs to connection %s\n",
+ adm_ctx.minor, adm_ctx.conn_name, adm_ctx.mdev->tconn->name);
+ drbd_msg_put_info("minor exists in different connection");
+ return ERR_INVALID_REQUEST;
+ }
+ if (adm_ctx.mdev &&
+ adm_ctx.volume != VOLUME_UNSPECIFIED &&
+ adm_ctx.volume != adm_ctx.mdev->vnr) {
+ pr_warning("request: minor=%u, volume=%u; but that minor is volume %u in %s\n",
+ adm_ctx.minor, adm_ctx.volume,
+ adm_ctx.mdev->vnr, adm_ctx.mdev->tconn->name);
+ drbd_msg_put_info("minor exists as different volume");
+ return ERR_INVALID_REQUEST;
+ }
+ if (adm_ctx.mdev && !adm_ctx.tconn)
+ adm_ctx.tconn = adm_ctx.mdev->tconn;
+ return NO_ERROR;
+
+fail:
+ nlmsg_free(adm_ctx.reply_skb);
+ adm_ctx.reply_skb = NULL;
+ return -ENOMEM;
+}
+
+static int drbd_adm_finish(struct genl_info *info, int retcode)
+{
+ struct nlattr *nla;
+ const char *conn_name = NULL;
+
+ if (!adm_ctx.reply_skb)
+ return -ENOMEM;
+
+ adm_ctx.reply_dh->ret_code = retcode;
+
+ nla = info->attrs[DRBD_NLA_CFG_CONTEXT];
+ if (nla) {
+ nla = nla_find_nested(nla, __nla_type(T_ctx_conn_name));
+ if (nla)
+ conn_name = nla_data(nla);
+ }
+
+ drbd_adm_send_reply(adm_ctx.reply_skb, info);
+ return 0;
+}
+
+static void setup_khelper_env(struct drbd_tconn *tconn, char **envp)
+{
+ char *afs;
+ struct net_conf *nc;
- if (get_net_conf(mdev->tconn)) {
- switch (((struct sockaddr *)mdev->tconn->net_conf->peer_addr)->sa_family) {
+ rcu_read_lock();
+ nc = rcu_dereference(tconn->net_conf);
+ if (nc) {
+ switch (((struct sockaddr *)nc->peer_addr)->sa_family) {
case AF_INET6:
afs = "ipv6";
- snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI6",
- &((struct sockaddr_in6 *)mdev->tconn->net_conf->peer_addr)->sin6_addr);
+ snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI6",
+ &((struct sockaddr_in6 *)nc->peer_addr)->sin6_addr);
break;
case AF_INET:
afs = "ipv4";
- snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI4",
- &((struct sockaddr_in *)mdev->tconn->net_conf->peer_addr)->sin_addr);
+ snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI4",
+ &((struct sockaddr_in *)nc->peer_addr)->sin_addr);
break;
default:
afs = "ssocks";
- snprintf(ad, 60, "DRBD_PEER_ADDRESS=%pI4",
- &((struct sockaddr_in *)mdev->tconn->net_conf->peer_addr)->sin_addr);
+ snprintf(envp[4], 60, "DRBD_PEER_ADDRESS=%pI4",
+ &((struct sockaddr_in *)nc->peer_addr)->sin_addr);
}
- snprintf(af, 20, "DRBD_PEER_AF=%s", afs);
- envp[3]=af;
- envp[4]=ad;
- put_net_conf(mdev->tconn);
+ snprintf(envp[3], 20, "DRBD_PEER_AF=%s", afs);
}
+ rcu_read_unlock();
+}
+
+int drbd_khelper(struct drbd_conf *mdev, char *cmd)
+{
+ char *envp[] = { "HOME=/",
+ "TERM=linux",
+ "PATH=/sbin:/usr/sbin:/bin:/usr/bin",
+ (char[20]) { }, /* address family */
+ (char[60]) { }, /* address */
+ NULL };
+ char mb[12];
+ char *argv[] = {usermode_helper, cmd, mb, NULL };
+ struct sib_info sib;
+ int ret;
+
+ snprintf(mb, 12, "minor-%d", mdev_to_minor(mdev));
+ setup_khelper_env(mdev->tconn, envp);
/* The helper may take some time.
* write out any unsynced meta data changes now */
drbd_md_sync(mdev);
dev_info(DEV, "helper command: %s %s %s\n", usermode_helper, cmd, mb);
-
- drbd_bcast_ev_helper(mdev, cmd);
+ sib.sib_reason = SIB_HELPER_PRE;
+ sib.helper_name = cmd;
+ drbd_bcast_event(mdev, &sib);
ret = call_usermodehelper(usermode_helper, argv, envp, 1);
if (ret)
dev_warn(DEV, "helper command: %s %s %s exit code %u (0x%x)\n",
dev_info(DEV, "helper command: %s %s %s exit code %u (0x%x)\n",
usermode_helper, cmd, mb,
(ret >> 8) & 0xff, ret);
+ sib.sib_reason = SIB_HELPER_POST;
+ sib.helper_exit_code = ret;
+ drbd_bcast_event(mdev, &sib);
+
+ if (ret < 0) /* Ignore any ERRNOs we got. */
+ ret = 0;
+
+ return ret;
+}
+
+static void conn_md_sync(struct drbd_tconn *tconn)
+{
+ struct drbd_conf *mdev;
+ int vnr;
+
+ down_read(&drbd_cfg_rwsem);
+ idr_for_each_entry(&tconn->volumes, mdev, vnr)
+ drbd_md_sync(mdev);
+ up_read(&drbd_cfg_rwsem);
+}
+
+int conn_khelper(struct drbd_tconn *tconn, char *cmd)
+{
+ char *envp[] = { "HOME=/",
+ "TERM=linux",
+ "PATH=/sbin:/usr/sbin:/bin:/usr/bin",
+ (char[20]) { }, /* address family */
+ (char[60]) { }, /* address */
+ NULL };
+ char *argv[] = {usermode_helper, cmd, tconn->name, NULL };
+ int ret;
+
+ setup_khelper_env(tconn, envp);
+ conn_md_sync(tconn);
+
+ conn_info(tconn, "helper command: %s %s %s\n", usermode_helper, cmd, tconn->name);
+ /* TODO: conn_bcast_event() ?? */
+
+ ret = call_usermodehelper(usermode_helper, argv, envp, 1);
+ if (ret)
+ conn_warn(tconn, "helper command: %s %s %s exit code %u (0x%x)\n",
+ usermode_helper, cmd, tconn->name,
+ (ret >> 8) & 0xff, ret);
+ else
+ conn_info(tconn, "helper command: %s %s %s exit code %u (0x%x)\n",
+ usermode_helper, cmd, tconn->name,
+ (ret >> 8) & 0xff, ret);
+ /* TODO: conn_bcast_event() ?? */
if (ret < 0) /* Ignore any ERRNOs we got. */
ret = 0;
return ret;
}
-enum drbd_disk_state drbd_try_outdate_peer(struct drbd_conf *mdev)
+static enum drbd_fencing_p highest_fencing_policy(struct drbd_tconn *tconn)
+{
+ enum drbd_fencing_p fp = FP_NOT_AVAIL;
+ struct drbd_conf *mdev;
+ int vnr;
+
+ rcu_read_lock();
+ idr_for_each_entry(&tconn->volumes, mdev, vnr) {
+ if (get_ldev_if_state(mdev, D_CONSISTENT)) {
+ fp = max_t(enum drbd_fencing_p, fp, mdev->ldev->dc.fencing);
+ put_ldev(mdev);
+ }
+ }
+ rcu_read_unlock();
+
+ return fp;
+}
+
+bool conn_try_outdate_peer(struct drbd_tconn *tconn)
{
+ union drbd_state mask = { };
+ union drbd_state val = { };
+ enum drbd_fencing_p fp;
char *ex_to_string;
int r;
- enum drbd_disk_state nps;
- enum drbd_fencing_p fp;
- D_ASSERT(mdev->state.pdsk == D_UNKNOWN);
+ if (tconn->cstate >= C_WF_REPORT_PARAMS) {
+ conn_err(tconn, "Expected cstate < C_WF_REPORT_PARAMS\n");
+ return false;
+ }
- if (get_ldev_if_state(mdev, D_CONSISTENT)) {
- fp = mdev->ldev->dc.fencing;
- put_ldev(mdev);
- } else {
- dev_warn(DEV, "Not fencing peer, I'm not even Consistent myself.\n");
- nps = mdev->state.pdsk;
+ fp = highest_fencing_policy(tconn);
+ switch (fp) {
+ case FP_NOT_AVAIL:
+ conn_warn(tconn, "Not fencing peer, I'm not even Consistent myself.\n");
goto out;
+ case FP_DONT_CARE:
+ return true;
+ default: ;
}
- r = drbd_khelper(mdev, "fence-peer");
+ r = conn_khelper(tconn, "fence-peer");
switch ((r>>8) & 0xff) {
case 3: /* peer is inconsistent */
ex_to_string = "peer is inconsistent or worse";
- nps = D_INCONSISTENT;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_INCONSISTENT;
break;
case 4: /* peer got outdated, or was already outdated */
ex_to_string = "peer was fenced";
- nps = D_OUTDATED;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_OUTDATED;
break;
case 5: /* peer was down */
- if (mdev->state.disk == D_UP_TO_DATE) {
+ if (conn_highest_disk(tconn) == D_UP_TO_DATE) {
/* we will(have) create(d) a new UUID anyways... */
ex_to_string = "peer is unreachable, assumed to be dead";
- nps = D_OUTDATED;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_OUTDATED;
} else {
ex_to_string = "peer unreachable, doing nothing since disk != UpToDate";
- nps = mdev->state.pdsk;
}
break;
case 6: /* Peer is primary, voluntarily outdate myself.
* This is useful when an unconnected R_SECONDARY is asked to
* become R_PRIMARY, but finds the other peer being active. */
ex_to_string = "peer is active";
- dev_warn(DEV, "Peer is primary, outdating myself.\n");
- nps = D_UNKNOWN;
- _drbd_request_state(mdev, NS(disk, D_OUTDATED), CS_WAIT_COMPLETE);
+ conn_warn(tconn, "Peer is primary, outdating myself.\n");
+ mask.disk = D_MASK;
+ val.disk = D_OUTDATED;
break;
case 7:
if (fp != FP_STONITH)
- dev_err(DEV, "fence-peer() = 7 && fencing != Stonith !!!\n");
+ conn_err(tconn, "fence-peer() = 7 && fencing != Stonith !!!\n");
ex_to_string = "peer was stonithed";
- nps = D_OUTDATED;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_OUTDATED;
break;
default:
/* The script is broken ... */
- nps = D_UNKNOWN;
- dev_err(DEV, "fence-peer helper broken, returned %d\n", (r>>8)&0xff);
- return nps;
+ conn_err(tconn, "fence-peer helper broken, returned %d\n", (r>>8)&0xff);
+ return false; /* Eventually leave IO frozen */
}
- dev_info(DEV, "fence-peer helper returned %d (%s)\n",
- (r>>8) & 0xff, ex_to_string);
+ conn_info(tconn, "fence-peer helper returned %d (%s)\n",
+ (r>>8) & 0xff, ex_to_string);
-out:
- if (mdev->state.susp_fen && nps >= D_UNKNOWN) {
- /* The handler was not successful... unfreeze here, the
- state engine can not unfreeze... */
- _drbd_request_state(mdev, NS(susp_fen, 0), CS_VERBOSE);
- }
+ out:
- return nps;
+ /* Not using
+ conn_request_state(tconn, mask, val, CS_VERBOSE);
+ here, because we might were able to re-establish the connection in the
+ meantime. */
+ spin_lock_irq(&tconn->req_lock);
+ if (tconn->cstate < C_WF_REPORT_PARAMS)
+ _conn_request_state(tconn, mask, val, CS_VERBOSE);
+ spin_unlock_irq(&tconn->req_lock);
+
+ return conn_highest_pdsk(tconn) <= D_OUTDATED;
}
static int _try_outdate_peer_async(void *data)
{
- struct drbd_conf *mdev = (struct drbd_conf *)data;
- enum drbd_disk_state nps;
- union drbd_state ns;
-
- nps = drbd_try_outdate_peer(mdev);
+ struct drbd_tconn *tconn = (struct drbd_tconn *)data;
- /* Not using
- drbd_request_state(mdev, NS(pdsk, nps));
- here, because we might were able to re-establish the connection
- in the meantime. This can only partially be solved in the state's
- engine is_valid_state() and is_valid_state_transition()
- functions.
-
- nps can be D_INCONSISTENT, D_OUTDATED or D_UNKNOWN.
- pdsk == D_INCONSISTENT while conn >= C_CONNECTED is valid,
- therefore we have to have the pre state change check here.
- */
- spin_lock_irq(&mdev->tconn->req_lock);
- ns = mdev->state;
- if (ns.conn < C_WF_REPORT_PARAMS) {
- ns.pdsk = nps;
- _drbd_set_state(mdev, ns, CS_VERBOSE, NULL);
- }
- spin_unlock_irq(&mdev->tconn->req_lock);
+ conn_try_outdate_peer(tconn);
+ kref_put(&tconn->kref, &conn_destroy);
return 0;
}
-void drbd_try_outdate_peer_async(struct drbd_conf *mdev)
+void conn_try_outdate_peer_async(struct drbd_tconn *tconn)
{
struct task_struct *opa;
- opa = kthread_run(_try_outdate_peer_async, mdev, "drbd%d_a_helper", mdev_to_minor(mdev));
- if (IS_ERR(opa))
- dev_err(DEV, "out of mem, failed to invoke fence-peer helper\n");
+ kref_get(&tconn->kref);
+ opa = kthread_run(_try_outdate_peer_async, tconn, "drbd_async_h");
+ if (IS_ERR(opa)) {
+ conn_err(tconn, "out of mem, failed to invoke fence-peer helper\n");
+ kref_put(&tconn->kref, &conn_destroy);
+ }
}
enum drbd_state_rv
{
const int max_tries = 4;
enum drbd_state_rv rv = SS_UNKNOWN_ERROR;
+ struct net_conf *nc;
int try = 0;
int forced = 0;
union drbd_state mask, val;
- enum drbd_disk_state nps;
if (new_role == R_PRIMARY)
- request_ping(mdev); /* Detect a dead peer ASAP */
+ request_ping(mdev->tconn); /* Detect a dead peer ASAP */
- mutex_lock(&mdev->state_mutex);
+ mutex_lock(mdev->state_mutex);
mask.i = 0; mask.role = R_MASK;
val.i = 0; val.role = new_role;
if (rv == SS_NO_UP_TO_DATE_DISK &&
mdev->state.disk == D_CONSISTENT && mask.pdsk == 0) {
D_ASSERT(mdev->state.pdsk == D_UNKNOWN);
- nps = drbd_try_outdate_peer(mdev);
- if (nps == D_OUTDATED || nps == D_INCONSISTENT) {
+ if (conn_try_outdate_peer(mdev->tconn)) {
val.disk = D_UP_TO_DATE;
mask.disk = D_MASK;
}
-
- val.pdsk = nps;
- mask.pdsk = D_MASK;
-
continue;
}
if (rv == SS_NOTHING_TO_DO)
- goto fail;
+ goto out;
if (rv == SS_PRIMARY_NOP && mask.pdsk == 0) {
- nps = drbd_try_outdate_peer(mdev);
-
- if (force && nps > D_OUTDATED) {
+ if (!conn_try_outdate_peer(mdev->tconn) && force) {
dev_warn(DEV, "Forced into split brain situation!\n");
- nps = D_OUTDATED;
- }
-
- mask.pdsk = D_MASK;
- val.pdsk = nps;
+ mask.pdsk = D_MASK;
+ val.pdsk = D_OUTDATED;
+ }
continue;
}
if (rv == SS_TWO_PRIMARIES) {
/* Maybe the peer is detected as dead very soon...
retry at most once more in this case. */
- schedule_timeout_interruptible((mdev->tconn->net_conf->ping_timeo+1)*HZ/10);
+ int timeo;
+ rcu_read_lock();
+ nc = rcu_dereference(mdev->tconn->net_conf);
+ timeo = nc ? (nc->ping_timeo + 1) * HZ / 10 : 1;
+ rcu_read_unlock();
+ schedule_timeout_interruptible(timeo);
if (try < max_tries)
try = max_tries - 1;
continue;
rv = _drbd_request_state(mdev, mask, val,
CS_VERBOSE + CS_WAIT_COMPLETE);
if (rv < SS_SUCCESS)
- goto fail;
+ goto out;
}
break;
}
if (rv < SS_SUCCESS)
- goto fail;
+ goto out;
if (forced)
dev_warn(DEV, "Forced to consider local data as UpToDate!\n");
put_ldev(mdev);
}
} else {
- if (get_net_conf(mdev->tconn)) {
- mdev->tconn->net_conf->want_lose = 0;
- put_net_conf(mdev->tconn);
- }
+ mutex_lock(&mdev->tconn->net_conf_update);
+ nc = mdev->tconn->net_conf;
+ if (nc)
+ nc->want_lose = 0; /* without copy; single bit op is atomic */
+ mutex_unlock(&mdev->tconn->net_conf_update);
+
set_disk_ro(mdev->vdisk, false);
if (get_ldev(mdev)) {
if (((mdev->state.conn < C_CONNECTED ||
drbd_md_sync(mdev);
kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE);
- fail:
- mutex_unlock(&mdev->state_mutex);
+out:
+ mutex_unlock(mdev->state_mutex);
return rv;
}
-static struct drbd_conf *ensure_mdev(int minor, int create)
+static const char *from_attrs_err_to_txt(int err)
{
- struct drbd_conf *mdev;
-
- if (minor >= minor_count)
- return NULL;
-
- mdev = minor_to_mdev(minor);
-
- if (!mdev && create) {
- struct gendisk *disk = NULL;
- mdev = drbd_new_device(minor);
-
- spin_lock_irq(&drbd_pp_lock);
- if (minor_table[minor] == NULL) {
- minor_table[minor] = mdev;
- disk = mdev->vdisk;
- mdev = NULL;
- } /* else: we lost the race */
- spin_unlock_irq(&drbd_pp_lock);
-
- if (disk) /* we won the race above */
- /* in case we ever add a drbd_delete_device(),
- * don't forget the del_gendisk! */
- add_disk(disk);
- else /* we lost the race above */
- drbd_free_mdev(mdev);
-
- mdev = minor_to_mdev(minor);
- }
-
- return mdev;
+ return err == -ENOMSG ? "required attribute missing" :
+ err == -EOPNOTSUPP ? "unknown mandatory attribute" :
+ err == -EEXIST ? "can not change invariant setting" :
+ "invalid attribute value";
}
-static int drbd_nl_primary(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_set_role(struct sk_buff *skb, struct genl_info *info)
{
- struct primary primary_args;
-
- memset(&primary_args, 0, sizeof(struct primary));
- if (!primary_from_tags(mdev, nlp->tag_list, &primary_args)) {
- reply->ret_code = ERR_MANDATORY_TAG;
- return 0;
- }
-
- reply->ret_code =
- drbd_set_role(mdev, R_PRIMARY, primary_args.primary_force);
+ struct set_role_parms parms;
+ int err;
+ enum drbd_ret_code retcode;
- return 0;
-}
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
-static int drbd_nl_secondary(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
-{
- reply->ret_code = drbd_set_role(mdev, R_SECONDARY, 0);
+ memset(&parms, 0, sizeof(parms));
+ if (info->attrs[DRBD_NLA_SET_ROLE_PARMS]) {
+ err = set_role_parms_from_attrs(&parms, info);
+ if (err) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto out;
+ }
+ }
+ if (info->genlhdr->cmd == DRBD_ADM_PRIMARY)
+ retcode = drbd_set_role(adm_ctx.mdev, R_PRIMARY, parms.assume_uptodate);
+ else
+ retcode = drbd_set_role(adm_ctx.mdev, R_SECONDARY, 0);
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
case DRBD_MD_INDEX_FLEX_INT:
bdev->md.md_offset = drbd_md_ss__(mdev, bdev);
/* al size is still fixed */
- bdev->md.al_offset = -MD_AL_MAX_SIZE;
+ bdev->md.al_offset = -MD_AL_SECTORS;
/* we need (slightly less than) ~ this much bitmap sectors: */
md_size_sect = drbd_get_capacity(bdev->backing_bdev);
md_size_sect = ALIGN(md_size_sect, BM_SECT_PER_EXT);
* R_PRIMARY D_INCONSISTENT, and C_SYNC_TARGET:
* peer may not initiate a resize.
*/
+/* Note these are not to be confused with
+ * drbd_adm_suspend_io/drbd_adm_resume_io,
+ * which are (sub) state changes triggered by admin (drbdsetup),
+ * and can be long lived.
+ * This changes an mdev->flag, is triggered by drbd internals,
+ * and should be short-lived. */
void drbd_suspend_io(struct drbd_conf *mdev)
{
set_bit(SUSPEND_IO, &mdev->flags);
- if (is_susp(mdev->state))
+ if (drbd_suspended(mdev))
return;
wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_bio_cnt));
}
* failed, and 0 on success. You should call drbd_md_sync() after you called
* this function.
*/
-static int drbd_check_al_size(struct drbd_conf *mdev)
+static int drbd_check_al_size(struct drbd_conf *mdev, struct disk_conf *dc)
{
struct lru_cache *n, *t;
struct lc_element *e;
unsigned int in_use;
int i;
- if (!expect(mdev->sync_conf.al_extents >= 7))
- mdev->sync_conf.al_extents = 127;
+ if (!expect(dc->al_extents >= DRBD_AL_EXTENTS_MIN))
+ dc->al_extents = DRBD_AL_EXTENTS_MIN;
if (mdev->act_log &&
- mdev->act_log->nr_elements == mdev->sync_conf.al_extents)
+ mdev->act_log->nr_elements == dc->al_extents)
return 0;
in_use = 0;
t = mdev->act_log;
- n = lc_create("act_log", drbd_al_ext_cache,
- mdev->sync_conf.al_extents, sizeof(struct lc_element), 0);
+ n = lc_create("act_log", drbd_al_ext_cache, AL_UPDATES_PER_TRANSACTION,
+ dc->al_extents, sizeof(struct lc_element), 0);
if (n == NULL) {
dev_err(DEV, "Cannot allocate act_log lru!\n");
* or start a new one. Flush any pending work, there may still be an
* after_state_change queued.
*/
-static void drbd_reconfig_start(struct drbd_conf *mdev)
+static void conn_reconfig_start(struct drbd_tconn *tconn)
{
- wait_event(mdev->state_wait, !test_and_set_bit(CONFIG_PENDING, &mdev->flags));
- wait_event(mdev->state_wait, !test_bit(DEVICE_DYING, &mdev->flags));
- drbd_thread_start(&mdev->tconn->worker);
- drbd_flush_workqueue(mdev->tconn);
+ wait_event(tconn->ping_wait, !test_and_set_bit(CONFIG_PENDING, &tconn->flags));
+ wait_event(tconn->ping_wait, !test_bit(OBJECT_DYING, &tconn->flags));
+ drbd_thread_start(&tconn->worker);
+ conn_flush_workqueue(tconn);
}
/* if still unconfigured, stops worker again.
* if configured now, clears CONFIG_PENDING.
* wakes potential waiters */
-static void drbd_reconfig_done(struct drbd_conf *mdev)
+static void conn_reconfig_done(struct drbd_tconn *tconn)
{
- spin_lock_irq(&mdev->tconn->req_lock);
- if (mdev->state.disk == D_DISKLESS &&
- mdev->state.conn == C_STANDALONE &&
- mdev->state.role == R_SECONDARY) {
- set_bit(DEVICE_DYING, &mdev->flags);
- drbd_thread_stop_nowait(&mdev->tconn->worker);
+ spin_lock_irq(&tconn->req_lock);
+ if (conn_all_vols_unconf(tconn)) {
+ set_bit(OBJECT_DYING, &tconn->flags);
+ drbd_thread_stop_nowait(&tconn->worker);
} else
- clear_bit(CONFIG_PENDING, &mdev->flags);
- spin_unlock_irq(&mdev->tconn->req_lock);
- wake_up(&mdev->state_wait);
+ clear_bit(CONFIG_PENDING, &tconn->flags);
+ spin_unlock_irq(&tconn->req_lock);
+ wake_up(&tconn->ping_wait);
}
/* Make sure IO is suspended before calling this function(). */
{
int s = 0;
- if (lc_try_lock(mdev->act_log)) {
- drbd_al_shrink(mdev);
- lc_unlock(mdev->act_log);
- } else {
+ if (!lc_try_lock(mdev->act_log)) {
dev_warn(DEV, "Failed to lock al in drbd_suspend_al()\n");
return;
}
+ drbd_al_shrink(mdev);
spin_lock_irq(&mdev->tconn->req_lock);
if (mdev->state.conn < C_CONNECTED)
s = !test_and_set_bit(AL_SUSPENDED, &mdev->flags);
-
spin_unlock_irq(&mdev->tconn->req_lock);
+ lc_unlock(mdev->act_log);
if (s)
dev_info(DEV, "Suspended AL updates\n");
}
-/* does always return 0;
- * interesting return code is in reply->ret_code */
-static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info)
+{
+ enum drbd_ret_code retcode;
+ struct drbd_conf *mdev;
+ struct disk_conf *ndc; /* new disk conf */
+ int err, fifo_size;
+ int *rs_plan_s = NULL;
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ mdev = adm_ctx.mdev;
+
+ /* we also need a disk
+ * to change the options on */
+ if (!get_ldev(mdev)) {
+ retcode = ERR_NO_DISK;
+ goto out;
+ }
+
+/* FIXME freeze IO, cluster wide.
+ *
+ * We should make sure no-one uses
+ * some half-updated struct when we
+ * assign it later. */
+
+ ndc = kmalloc(sizeof(*ndc), GFP_KERNEL);
+ if (!ndc) {
+ retcode = ERR_NOMEM;
+ goto fail;
+ }
+
+ memcpy(ndc, &mdev->ldev->dc, sizeof(*ndc));
+ err = disk_conf_from_attrs_for_change(ndc, info);
+ if (err) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ }
+
+ if (!expect(ndc->resync_rate >= 1))
+ ndc->resync_rate = 1;
+
+ /* clip to allowed range */
+ if (!expect(ndc->al_extents >= DRBD_AL_EXTENTS_MIN))
+ ndc->al_extents = DRBD_AL_EXTENTS_MIN;
+ if (!expect(ndc->al_extents <= DRBD_AL_EXTENTS_MAX))
+ ndc->al_extents = DRBD_AL_EXTENTS_MAX;
+
+ /* most sanity checks done, try to assign the new sync-after
+ * dependency. need to hold the global lock in there,
+ * to avoid a race in the dependency loop check. */
+ retcode = drbd_alter_sa(mdev, ndc->resync_after);
+ if (retcode != NO_ERROR)
+ goto fail;
+
+ fifo_size = (ndc->c_plan_ahead * 10 * SLEEP_TIME) / HZ;
+ if (fifo_size != mdev->rs_plan_s.size && fifo_size > 0) {
+ rs_plan_s = kzalloc(sizeof(int) * fifo_size, GFP_KERNEL);
+ if (!rs_plan_s) {
+ dev_err(DEV, "kmalloc of fifo_buffer failed");
+ retcode = ERR_NOMEM;
+ goto fail;
+ }
+ }
+
+ if (fifo_size != mdev->rs_plan_s.size) {
+ kfree(mdev->rs_plan_s.values);
+ mdev->rs_plan_s.values = rs_plan_s;
+ mdev->rs_plan_s.size = fifo_size;
+ mdev->rs_planed = 0;
+ rs_plan_s = NULL;
+ }
+
+ wait_event(mdev->al_wait, lc_try_lock(mdev->act_log));
+ drbd_al_shrink(mdev);
+ err = drbd_check_al_size(mdev, ndc);
+ lc_unlock(mdev->act_log);
+ wake_up(&mdev->al_wait);
+
+ if (err) {
+ retcode = ERR_NOMEM;
+ goto fail;
+ }
+
+ /* FIXME
+ * To avoid someone looking at a half-updated struct, we probably
+ * should have a rw-semaphor on net_conf and disk_conf.
+ */
+ mdev->ldev->dc = *ndc;
+
+ drbd_md_sync(mdev);
+
+
+ if (mdev->state.conn >= C_CONNECTED)
+ drbd_send_sync_param(mdev);
+
+ fail:
+ put_ldev(mdev);
+ kfree(ndc);
+ kfree(rs_plan_s);
+ out:
+ drbd_adm_finish(info, retcode);
+ return 0;
+}
+
+int drbd_adm_attach(struct sk_buff *skb, struct genl_info *info)
{
+ struct drbd_conf *mdev;
+ int err;
enum drbd_ret_code retcode;
enum determine_dev_size dd;
sector_t max_possible_sectors;
struct lru_cache *resync_lru = NULL;
union drbd_state ns, os;
enum drbd_state_rv rv;
+ struct net_conf *nc;
int cp_discovered = 0;
- int logical_block_size;
- drbd_reconfig_start(mdev);
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto finish;
+
+ mdev = adm_ctx.mdev;
+ conn_reconfig_start(mdev->tconn);
/* if you want to reconfigure, please tear down first */
if (mdev->state.disk > D_DISKLESS) {
* to realize a "hot spare" feature (not that I'd recommend that) */
wait_event(mdev->misc_wait, !atomic_read(&mdev->local_cnt));
- /* allocation not in the IO path, cqueue thread context */
+ /* allocation not in the IO path, drbdsetup context */
nbc = kzalloc(sizeof(struct drbd_backing_dev), GFP_KERNEL);
if (!nbc) {
retcode = ERR_NOMEM;
goto fail;
}
- nbc->dc.disk_size = DRBD_DISK_SIZE_SECT_DEF;
- nbc->dc.on_io_error = DRBD_ON_IO_ERROR_DEF;
- nbc->dc.fencing = DRBD_FENCING_DEF;
- nbc->dc.max_bio_bvecs = DRBD_MAX_BIO_BVECS_DEF;
-
- if (!disk_conf_from_tags(mdev, nlp->tag_list, &nbc->dc)) {
+ nbc->dc = (struct disk_conf) {
+ {}, 0, /* backing_dev */
+ {}, 0, /* meta_dev */
+ 0, /* meta_dev_idx */
+ DRBD_DISK_SIZE_SECT_DEF, /* disk_size */
+ DRBD_MAX_BIO_BVECS_DEF, /* max_bio_bvecs */
+ DRBD_ON_IO_ERROR_DEF, /* on_io_error */
+ DRBD_FENCING_DEF, /* fencing */
+ DRBD_RATE_DEF, /* resync_rate */
+ DRBD_AFTER_DEF, /* resync_after */
+ DRBD_AL_EXTENTS_DEF, /* al_extents */
+ DRBD_C_PLAN_AHEAD_DEF, /* c_plan_ahead */
+ DRBD_C_DELAY_TARGET_DEF, /* c_delay_target */
+ DRBD_C_FILL_TARGET_DEF, /* c_fill_target */
+ DRBD_C_MAX_RATE_DEF, /* c_max_rate */
+ DRBD_C_MIN_RATE_DEF, /* c_min_rate */
+ 0, /* no_disk_barrier */
+ 0, /* no_disk_flush */
+ 0, /* no_disk_drain */
+ 0, /* no_md_flush */
+ };
+
+ err = disk_conf_from_attrs(&nbc->dc, info);
+ if (err) {
retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
goto fail;
}
- if (nbc->dc.meta_dev_idx < DRBD_MD_INDEX_FLEX_INT) {
+ if ((int)nbc->dc.meta_dev_idx < DRBD_MD_INDEX_FLEX_INT) {
retcode = ERR_MD_IDX_INVALID;
goto fail;
}
- if (get_net_conf(mdev->tconn)) {
- int prot = mdev->tconn->net_conf->wire_protocol;
- put_net_conf(mdev->tconn);
- if (nbc->dc.fencing == FP_STONITH && prot == DRBD_PROT_A) {
+ rcu_read_lock();
+ nc = rcu_dereference(mdev->tconn->net_conf);
+ if (nc) {
+ if (nbc->dc.fencing == FP_STONITH && nc->wire_protocol == DRBD_PROT_A) {
+ rcu_read_unlock();
retcode = ERR_STONITH_AND_PROT_A;
goto fail;
}
}
+ rcu_read_unlock();
bdev = blkdev_get_by_path(nbc->dc.backing_dev,
FMODE_READ | FMODE_WRITE | FMODE_EXCL, mdev);
*/
bdev = blkdev_get_by_path(nbc->dc.meta_dev,
FMODE_READ | FMODE_WRITE | FMODE_EXCL,
- (nbc->dc.meta_dev_idx < 0) ?
+ ((int)nbc->dc.meta_dev_idx < 0) ?
(void *)mdev : (void *)drbd_m_holder);
if (IS_ERR(bdev)) {
dev_err(DEV, "open(\"%s\") failed with %ld\n", nbc->dc.meta_dev,
}
resync_lru = lc_create("resync", drbd_bm_ext_cache,
- 61, sizeof(struct bm_extent),
+ 1, 61, sizeof(struct bm_extent),
offsetof(struct bm_extent, lce));
if (!resync_lru) {
retcode = ERR_NOMEM;
goto fail;
}
- if (nbc->dc.meta_dev_idx < 0) {
+ if ((int)nbc->dc.meta_dev_idx < 0) {
max_possible_sectors = DRBD_MAX_SECTORS_FLEX;
/* at least one MB, otherwise it does not make sense */
min_md_device_sectors = (2<<10);
dev_warn(DEV, "==> truncating very big lower level device "
"to currently maximum possible %llu sectors <==\n",
(unsigned long long) max_possible_sectors);
- if (nbc->dc.meta_dev_idx >= 0)
+ if ((int)nbc->dc.meta_dev_idx >= 0)
dev_warn(DEV, "==>> using internal or flexible "
"meta data may help <<==\n");
}
drbd_suspend_io(mdev);
/* also wait for the last barrier ack. */
- wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt) || is_susp(mdev->state));
+ wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt) || drbd_suspended(mdev));
/* and for any other previously queued work */
- drbd_flush_workqueue(mdev->tconn);
+ drbd_flush_workqueue(mdev);
rv = _drbd_request_state(mdev, NS(disk, D_ATTACHING), CS_VERBOSE);
retcode = rv; /* FIXME: Type mismatch. */
drbd_md_set_sector_offsets(mdev, nbc);
- /* allocate a second IO page if logical_block_size != 512 */
- logical_block_size = bdev_logical_block_size(nbc->md_bdev);
- if (logical_block_size == 0)
- logical_block_size = MD_SECTOR_SIZE;
-
- if (logical_block_size != MD_SECTOR_SIZE) {
- if (!mdev->md_io_tmpp) {
- struct page *page = alloc_page(GFP_NOIO);
- if (!page)
- goto force_diskless_dec;
-
- dev_warn(DEV, "Meta data's bdev logical_block_size = %d != %d\n",
- logical_block_size, MD_SECTOR_SIZE);
- dev_warn(DEV, "Workaround engaged (has performance impact).\n");
-
- mdev->md_io_tmpp = page;
- }
- }
-
if (!mdev->bitmap) {
if (drbd_bm_init(mdev)) {
retcode = ERR_NOMEM;
}
/* Since we are diskless, fix the activity log first... */
- if (drbd_check_al_size(mdev)) {
+ if (drbd_check_al_size(mdev, &nbc->dc)) {
retcode = ERR_NOMEM;
goto force_diskless_dec;
}
clear_bit(CRASHED_PRIMARY, &mdev->flags);
if (drbd_md_test_flag(mdev->ldev, MDF_PRIMARY_IND) &&
- !(mdev->state.role == R_PRIMARY && mdev->state.susp_nod)) {
+ !(mdev->state.role == R_PRIMARY && mdev->tconn->susp_nod)) {
set_bit(CRASHED_PRIMARY, &mdev->flags);
cp_discovered = 1;
}
}
} else {
if (drbd_bitmap_io(mdev, &drbd_bm_read,
- "read from attaching", BM_LOCKED_MASK) < 0) {
+ "read from attaching", BM_LOCKED_MASK)) {
retcode = ERR_IO_MD_DISK;
goto force_diskless_dec;
}
drbd_suspend_al(mdev); /* IO is still suspended here... */
spin_lock_irq(&mdev->tconn->req_lock);
- os = mdev->state;
- ns.i = os.i;
+ os = drbd_read_state(mdev);
+ ns = os;
/* If MDF_CONSISTENT is not set go into inconsistent state,
otherwise investigate MDF_WasUpToDate...
If MDF_WAS_UP_TO_DATE is not set go into D_OUTDATED disk state,
}
rv = _drbd_set_state(mdev, ns, CS_VERBOSE, NULL);
- ns = mdev->state;
spin_unlock_irq(&mdev->tconn->req_lock);
if (rv < SS_SUCCESS)
kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE);
put_ldev(mdev);
- reply->ret_code = retcode;
- drbd_reconfig_done(mdev);
+ conn_reconfig_done(mdev->tconn);
+ drbd_adm_finish(info, retcode);
return 0;
force_diskless_dec:
drbd_force_state(mdev, NS(disk, D_FAILED));
drbd_md_sync(mdev);
fail:
+ conn_reconfig_done(mdev->tconn);
if (nbc) {
if (nbc->backing_bdev)
blkdev_put(nbc->backing_bdev,
}
lc_destroy(resync_lru);
- reply->ret_code = retcode;
- drbd_reconfig_done(mdev);
+ finish:
+ drbd_adm_finish(info, retcode);
return 0;
}
-/* Detaching the disk is a process in multiple stages. First we need to lock
- * out application IO, in-flight IO, IO stuck in drbd_al_begin_io.
- * Then we transition to D_DISKLESS, and wait for put_ldev() to return all
+static int adm_detach(struct drbd_conf *mdev)
+{
+ enum drbd_state_rv retcode;
+ drbd_suspend_io(mdev); /* so no-one is stuck in drbd_al_begin_io */
+ retcode = drbd_request_state(mdev, NS(disk, D_DISKLESS));
+ wait_event(mdev->misc_wait,
+ mdev->state.disk != D_DISKLESS ||
+ !atomic_read(&mdev->local_cnt));
+ drbd_resume_io(mdev);
+ return retcode;
+}
+
+/* Detaching the disk is a process in multiple stages. First we need to lock
+ * out application IO, in-flight IO, IO stuck in drbd_al_begin_io.
+ * Then we transition to D_DISKLESS, and wait for put_ldev() to return all
* internal references as well.
* Only then we have finally detached. */
-static int drbd_nl_detach(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_detach(struct sk_buff *skb, struct genl_info *info)
{
enum drbd_ret_code retcode;
- int ret;
- drbd_suspend_io(mdev); /* so no-one is stuck in drbd_al_begin_io */
- retcode = drbd_request_state(mdev, NS(disk, D_FAILED));
- /* D_FAILED will transition to DISKLESS. */
- ret = wait_event_interruptible(mdev->misc_wait,
- mdev->state.disk != D_FAILED);
- drbd_resume_io(mdev);
- if ((int)retcode == (int)SS_IS_DISKLESS)
- retcode = SS_NOTHING_TO_DO;
- if (ret)
- retcode = ERR_INTR;
- reply->ret_code = retcode;
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ retcode = adm_detach(adm_ctx.mdev);
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_net_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+static bool conn_resync_running(struct drbd_tconn *tconn)
{
+ struct drbd_conf *mdev;
+ bool rv = false;
+ int vnr;
+
+ rcu_read_lock();
+ idr_for_each_entry(&tconn->volumes, mdev, vnr) {
+ if (mdev->state.conn == C_SYNC_SOURCE ||
+ mdev->state.conn == C_SYNC_TARGET ||
+ mdev->state.conn == C_PAUSED_SYNC_S ||
+ mdev->state.conn == C_PAUSED_SYNC_T) {
+ rv = true;
+ break;
+ }
+ }
+ rcu_read_unlock();
+
+ return rv;
+}
+
+static bool conn_ov_running(struct drbd_tconn *tconn)
+{
+ struct drbd_conf *mdev;
+ bool rv = false;
+ int vnr;
+
+ rcu_read_lock();
+ idr_for_each_entry(&tconn->volumes, mdev, vnr) {
+ if (mdev->state.conn == C_VERIFY_S ||
+ mdev->state.conn == C_VERIFY_T) {
+ rv = true;
+ break;
+ }
+ }
+ rcu_read_unlock();
+
+ return rv;
+}
+
+static enum drbd_ret_code
+_check_net_options(struct drbd_tconn *tconn, struct net_conf *old_conf, struct net_conf *new_conf)
+{
+ struct drbd_conf *mdev;
int i;
- enum drbd_ret_code retcode;
- struct net_conf *new_conf = NULL;
- struct crypto_hash *tfm = NULL;
- struct crypto_hash *integrity_w_tfm = NULL;
- struct crypto_hash *integrity_r_tfm = NULL;
- struct drbd_conf *odev;
- char hmac_name[CRYPTO_MAX_ALG_NAME];
- void *int_dig_out = NULL;
- void *int_dig_in = NULL;
- void *int_dig_vv = NULL;
- struct sockaddr *new_my_addr, *new_peer_addr, *taken_addr;
- drbd_reconfig_start(mdev);
+ if (old_conf && tconn->agreed_pro_version < 100 &&
+ tconn->cstate == C_WF_REPORT_PARAMS &&
+ new_conf->wire_protocol != old_conf->wire_protocol)
+ return ERR_NEED_APV_100;
- if (mdev->state.conn > C_STANDALONE) {
- retcode = ERR_NET_CONFIGURED;
- goto fail;
+ if (new_conf->two_primaries &&
+ (new_conf->wire_protocol != DRBD_PROT_C))
+ return ERR_NOT_PROTO_C;
+
+ idr_for_each_entry(&tconn->volumes, mdev, i) {
+ if (get_ldev(mdev)) {
+ enum drbd_fencing_p fp = mdev->ldev->dc.fencing;
+ put_ldev(mdev);
+ if (new_conf->wire_protocol == DRBD_PROT_A && fp == FP_STONITH)
+ return ERR_STONITH_AND_PROT_A;
+ }
+ if (mdev->state.role == R_PRIMARY && new_conf->want_lose)
+ return ERR_DISCARD;
}
- /* allocation not in the IO path, cqueue thread context */
+ if (new_conf->on_congestion != OC_BLOCK && new_conf->wire_protocol != DRBD_PROT_A)
+ return ERR_CONG_NOT_PROTO_A;
+
+ return NO_ERROR;
+}
+
+static enum drbd_ret_code
+check_net_options(struct drbd_tconn *tconn, struct net_conf *new_conf)
+{
+ static enum drbd_ret_code rv;
+ struct drbd_conf *mdev;
+ int i;
+
+ rcu_read_lock();
+ rv = _check_net_options(tconn, rcu_dereference(tconn->net_conf), new_conf);
+ rcu_read_unlock();
+
+ /* tconn->volumes protected by genl_lock() here */
+ idr_for_each_entry(&tconn->volumes, mdev, i) {
+ if (!mdev->bitmap) {
+ if(drbd_bm_init(mdev))
+ return ERR_NOMEM;
+ }
+ }
+
+ return rv;
+}
+
+int drbd_adm_net_opts(struct sk_buff *skb, struct genl_info *info)
+{
+ enum drbd_ret_code retcode;
+ struct drbd_tconn *tconn;
+ struct net_conf *old_conf, *new_conf = NULL;
+ int err;
+ int ovr; /* online verify running */
+ int rsr; /* re-sync running */
+ struct crypto_hash *verify_tfm = NULL;
+ struct crypto_hash *csums_tfm = NULL;
+
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ tconn = adm_ctx.tconn;
+
new_conf = kzalloc(sizeof(struct net_conf), GFP_KERNEL);
if (!new_conf) {
retcode = ERR_NOMEM;
+ goto out;
+ }
+
+ conn_reconfig_start(tconn);
+
+ mutex_lock(&tconn->net_conf_update);
+ old_conf = tconn->net_conf;
+
+ if (!old_conf) {
+ drbd_msg_put_info("net conf missing, try connect");
+ retcode = ERR_INVALID_REQUEST;
goto fail;
}
- new_conf->timeout = DRBD_TIMEOUT_DEF;
- new_conf->try_connect_int = DRBD_CONNECT_INT_DEF;
- new_conf->ping_int = DRBD_PING_INT_DEF;
- new_conf->max_epoch_size = DRBD_MAX_EPOCH_SIZE_DEF;
- new_conf->max_buffers = DRBD_MAX_BUFFERS_DEF;
- new_conf->unplug_watermark = DRBD_UNPLUG_WATERMARK_DEF;
- new_conf->sndbuf_size = DRBD_SNDBUF_SIZE_DEF;
- new_conf->rcvbuf_size = DRBD_RCVBUF_SIZE_DEF;
- new_conf->ko_count = DRBD_KO_COUNT_DEF;
- new_conf->after_sb_0p = DRBD_AFTER_SB_0P_DEF;
- new_conf->after_sb_1p = DRBD_AFTER_SB_1P_DEF;
- new_conf->after_sb_2p = DRBD_AFTER_SB_2P_DEF;
- new_conf->want_lose = 0;
- new_conf->two_primaries = 0;
- new_conf->wire_protocol = DRBD_PROT_C;
- new_conf->ping_timeo = DRBD_PING_TIMEO_DEF;
- new_conf->rr_conflict = DRBD_RR_CONFLICT_DEF;
- new_conf->on_congestion = DRBD_ON_CONGESTION_DEF;
- new_conf->cong_extents = DRBD_CONG_EXTENTS_DEF;
-
- if (!net_conf_from_tags(mdev, nlp->tag_list, new_conf)) {
+ *new_conf = *old_conf;
+
+ err = net_conf_from_attrs_for_change(new_conf, info);
+ if (err) {
retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
goto fail;
}
- if (new_conf->two_primaries
- && (new_conf->wire_protocol != DRBD_PROT_C)) {
- retcode = ERR_NOT_PROTO_C;
+ retcode = check_net_options(tconn, new_conf);
+ if (retcode != NO_ERROR)
+ goto fail;
+
+ /* re-sync running */
+ rsr = conn_resync_running(tconn);
+ if (rsr && old_conf && strcmp(new_conf->csums_alg, old_conf->csums_alg)) {
+ retcode = ERR_CSUMS_RESYNC_RUNNING;
goto fail;
}
- if (get_ldev(mdev)) {
- enum drbd_fencing_p fp = mdev->ldev->dc.fencing;
- put_ldev(mdev);
- if (new_conf->wire_protocol == DRBD_PROT_A && fp == FP_STONITH) {
- retcode = ERR_STONITH_AND_PROT_A;
+ if (!rsr && new_conf->csums_alg[0]) {
+ csums_tfm = crypto_alloc_hash(new_conf->csums_alg, 0, CRYPTO_ALG_ASYNC);
+ if (IS_ERR(csums_tfm)) {
+ csums_tfm = NULL;
+ retcode = ERR_CSUMS_ALG;
+ goto fail;
+ }
+
+ if (!drbd_crypto_is_hash(crypto_hash_tfm(csums_tfm))) {
+ retcode = ERR_CSUMS_ALG_ND;
goto fail;
}
}
- if (new_conf->on_congestion != OC_BLOCK && new_conf->wire_protocol != DRBD_PROT_A) {
- retcode = ERR_CONG_NOT_PROTO_A;
+ /* online verify running */
+ ovr = conn_ov_running(tconn);
+ if (ovr) {
+ if (strcmp(new_conf->verify_alg, old_conf->verify_alg)) {
+ retcode = ERR_VERIFY_RUNNING;
+ goto fail;
+ }
+ }
+
+ if (!ovr && new_conf->verify_alg[0]) {
+ verify_tfm = crypto_alloc_hash(new_conf->verify_alg, 0, CRYPTO_ALG_ASYNC);
+ if (IS_ERR(verify_tfm)) {
+ verify_tfm = NULL;
+ retcode = ERR_VERIFY_ALG;
+ goto fail;
+ }
+
+ if (!drbd_crypto_is_hash(crypto_hash_tfm(verify_tfm))) {
+ retcode = ERR_VERIFY_ALG_ND;
+ goto fail;
+ }
+ }
+
+ rcu_assign_pointer(tconn->net_conf, new_conf);
+
+ if (!rsr) {
+ crypto_free_hash(tconn->csums_tfm);
+ tconn->csums_tfm = csums_tfm;
+ csums_tfm = NULL;
+ }
+ if (!ovr) {
+ crypto_free_hash(tconn->verify_tfm);
+ tconn->verify_tfm = verify_tfm;
+ verify_tfm = NULL;
+ }
+
+ mutex_unlock(&tconn->net_conf_update);
+ synchronize_rcu();
+ kfree(old_conf);
+
+ if (tconn->cstate >= C_WF_REPORT_PARAMS)
+ drbd_send_sync_param(minor_to_mdev(conn_lowest_minor(tconn)));
+
+ goto done;
+
+ fail:
+ mutex_unlock(&tconn->net_conf_update);
+ crypto_free_hash(csums_tfm);
+ crypto_free_hash(verify_tfm);
+ kfree(new_conf);
+ done:
+ conn_reconfig_done(tconn);
+ out:
+ drbd_adm_finish(info, retcode);
+ return 0;
+}
+
+int drbd_adm_connect(struct sk_buff *skb, struct genl_info *info)
+{
+ char hmac_name[CRYPTO_MAX_ALG_NAME];
+ struct drbd_conf *mdev;
+ struct net_conf *old_conf, *new_conf = NULL;
+ struct crypto_hash *tfm = NULL;
+ struct crypto_hash *integrity_w_tfm = NULL;
+ struct crypto_hash *integrity_r_tfm = NULL;
+ void *int_dig_in = NULL;
+ void *int_dig_vv = NULL;
+ struct drbd_tconn *oconn;
+ struct drbd_tconn *tconn;
+ struct sockaddr *new_my_addr, *new_peer_addr, *taken_addr;
+ enum drbd_ret_code retcode;
+ int i;
+ int err;
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ tconn = adm_ctx.tconn;
+ conn_reconfig_start(tconn);
+
+ if (tconn->cstate > C_STANDALONE) {
+ retcode = ERR_NET_CONFIGURED;
goto fail;
}
- if (mdev->state.role == R_PRIMARY && new_conf->want_lose) {
- retcode = ERR_DISCARD;
+ /* allocation not in the IO path, cqueue thread context */
+ new_conf = kmalloc(sizeof(struct net_conf), GFP_KERNEL);
+ if (!new_conf) {
+ retcode = ERR_NOMEM;
goto fail;
}
+ *new_conf = (struct net_conf) {
+ {}, 0, /* my_addr */
+ {}, 0, /* peer_addr */
+ {}, 0, /* shared_secret */
+ {}, 0, /* cram_hmac_alg */
+ {}, 0, /* integrity_alg */
+ {}, 0, /* verify_alg */
+ {}, 0, /* csums_alg */
+ DRBD_PROTOCOL_DEF, /* wire_protocol */
+ DRBD_CONNECT_INT_DEF, /* try_connect_int */
+ DRBD_TIMEOUT_DEF, /* timeout */
+ DRBD_PING_INT_DEF, /* ping_int */
+ DRBD_PING_TIMEO_DEF, /* ping_timeo */
+ DRBD_SNDBUF_SIZE_DEF, /* sndbuf_size */
+ DRBD_RCVBUF_SIZE_DEF, /* rcvbuf_size */
+ DRBD_KO_COUNT_DEF, /* ko_count */
+ DRBD_MAX_BUFFERS_DEF, /* max_buffers */
+ DRBD_MAX_EPOCH_SIZE_DEF, /* max_epoch_size */
+ DRBD_UNPLUG_WATERMARK_DEF, /* unplug_watermark */
+ DRBD_AFTER_SB_0P_DEF, /* after_sb_0p */
+ DRBD_AFTER_SB_1P_DEF, /* after_sb_1p */
+ DRBD_AFTER_SB_2P_DEF, /* after_sb_2p */
+ DRBD_RR_CONFLICT_DEF, /* rr_conflict */
+ DRBD_ON_CONGESTION_DEF, /* on_congestion */
+ DRBD_CONG_FILL_DEF, /* cong_fill */
+ DRBD_CONG_EXTENTS_DEF, /* cong_extents */
+ 0, /* two_primaries */
+ 0, /* want_lose */
+ 0, /* no_cork */
+ 0, /* always_asbp */
+ 0, /* dry_run */
+ 0, /* use_rle */
+ };
+
+ err = net_conf_from_attrs(new_conf, info);
+ if (err) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto fail;
+ }
+
+ retcode = check_net_options(tconn, new_conf);
+ if (retcode != NO_ERROR)
+ goto fail;
+
retcode = NO_ERROR;
new_my_addr = (struct sockaddr *)&new_conf->my_addr;
new_peer_addr = (struct sockaddr *)&new_conf->peer_addr;
- for (i = 0; i < minor_count; i++) {
- odev = minor_to_mdev(i);
- if (!odev || odev == mdev)
+
+ /* No need to take drbd_cfg_rwsem here. All reconfiguration is
+ * strictly serialized on genl_lock(). We are protected against
+ * concurrent reconfiguration/addition/deletion */
+ list_for_each_entry(oconn, &drbd_tconns, all_tconn) {
+ struct net_conf *nc;
+ if (oconn == tconn)
continue;
- if (get_net_conf(odev->tconn)) {
- taken_addr = (struct sockaddr *)&odev->tconn->net_conf->my_addr;
- if (new_conf->my_addr_len == odev->tconn->net_conf->my_addr_len &&
+
+ rcu_read_lock();
+ nc = rcu_dereference(oconn->net_conf);
+ if (nc) {
+ taken_addr = (struct sockaddr *)&nc->my_addr;
+ if (new_conf->my_addr_len == nc->my_addr_len &&
!memcmp(new_my_addr, taken_addr, new_conf->my_addr_len))
retcode = ERR_LOCAL_ADDR;
- taken_addr = (struct sockaddr *)&odev->tconn->net_conf->peer_addr;
- if (new_conf->peer_addr_len == odev->tconn->net_conf->peer_addr_len &&
+ taken_addr = (struct sockaddr *)&nc->peer_addr;
+ if (new_conf->peer_addr_len == nc->peer_addr_len &&
!memcmp(new_peer_addr, taken_addr, new_conf->peer_addr_len))
retcode = ERR_PEER_ADDR;
-
- put_net_conf(odev->tconn);
- if (retcode != NO_ERROR)
- goto fail;
}
+ rcu_read_unlock();
+ if (retcode != NO_ERROR)
+ goto fail;
}
if (new_conf->cram_hmac_alg[0] != 0) {
((char *)new_conf->shared_secret)[SHARED_SECRET_MAX-1] = 0;
+ /* allocation not in the IO path, cqueue thread context */
if (integrity_w_tfm) {
i = crypto_hash_digestsize(integrity_w_tfm);
- int_dig_out = kmalloc(i, GFP_KERNEL);
- if (!int_dig_out) {
- retcode = ERR_NOMEM;
- goto fail;
- }
int_dig_in = kmalloc(i, GFP_KERNEL);
if (!int_dig_in) {
retcode = ERR_NOMEM;
}
}
- if (!mdev->bitmap) {
- if(drbd_bm_init(mdev)) {
- retcode = ERR_NOMEM;
- goto fail;
- }
- }
+ conn_flush_workqueue(tconn);
- drbd_flush_workqueue(mdev->tconn);
- spin_lock_irq(&mdev->tconn->req_lock);
- if (mdev->tconn->net_conf != NULL) {
+ mutex_lock(&tconn->net_conf_update);
+ old_conf = tconn->net_conf;
+ if (old_conf) {
retcode = ERR_NET_CONFIGURED;
- spin_unlock_irq(&mdev->tconn->req_lock);
+ mutex_unlock(&tconn->net_conf_update);
goto fail;
}
- mdev->tconn->net_conf = new_conf;
+ rcu_assign_pointer(tconn->net_conf, new_conf);
- mdev->send_cnt = 0;
- mdev->recv_cnt = 0;
-
- crypto_free_hash(mdev->tconn->cram_hmac_tfm);
- mdev->tconn->cram_hmac_tfm = tfm;
+ conn_free_crypto(tconn);
+ tconn->cram_hmac_tfm = tfm;
+ tconn->integrity_w_tfm = integrity_w_tfm;
+ tconn->integrity_r_tfm = integrity_r_tfm;
+ tconn->int_dig_in = int_dig_in;
+ tconn->int_dig_vv = int_dig_vv;
- crypto_free_hash(mdev->tconn->integrity_w_tfm);
- mdev->tconn->integrity_w_tfm = integrity_w_tfm;
+ mutex_unlock(&tconn->net_conf_update);
- crypto_free_hash(mdev->tconn->integrity_r_tfm);
- mdev->tconn->integrity_r_tfm = integrity_r_tfm;
+ retcode = conn_request_state(tconn, NS(conn, C_UNCONNECTED), CS_VERBOSE);
- kfree(mdev->tconn->int_dig_out);
- kfree(mdev->tconn->int_dig_in);
- kfree(mdev->tconn->int_dig_vv);
- mdev->tconn->int_dig_out=int_dig_out;
- mdev->tconn->int_dig_in=int_dig_in;
- mdev->tconn->int_dig_vv=int_dig_vv;
- retcode = _drbd_set_state(_NS(mdev, conn, C_UNCONNECTED), CS_VERBOSE, NULL);
- spin_unlock_irq(&mdev->tconn->req_lock);
-
- kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE);
- reply->ret_code = retcode;
- drbd_reconfig_done(mdev);
+ rcu_read_lock();
+ idr_for_each_entry(&tconn->volumes, mdev, i) {
+ mdev->send_cnt = 0;
+ mdev->recv_cnt = 0;
+ kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE);
+ }
+ rcu_read_unlock();
+ conn_reconfig_done(tconn);
+ drbd_adm_finish(info, retcode);
return 0;
fail:
- kfree(int_dig_out);
kfree(int_dig_in);
kfree(int_dig_vv);
crypto_free_hash(tfm);
crypto_free_hash(integrity_r_tfm);
kfree(new_conf);
- reply->ret_code = retcode;
- drbd_reconfig_done(mdev);
+ conn_reconfig_done(tconn);
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_disconnect(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+static enum drbd_state_rv conn_try_disconnect(struct drbd_tconn *tconn, bool force)
{
- int retcode;
- struct disconnect dc;
-
- memset(&dc, 0, sizeof(struct disconnect));
- if (!disconnect_from_tags(mdev, nlp->tag_list, &dc)) {
- retcode = ERR_MANDATORY_TAG;
- goto fail;
+ enum drbd_state_rv rv;
+ if (force) {
+ spin_lock_irq(&tconn->req_lock);
+ if (tconn->cstate >= C_WF_CONNECTION)
+ _conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD);
+ spin_unlock_irq(&tconn->req_lock);
+ return SS_SUCCESS;
+ }
+
+ rv = conn_request_state(tconn, NS(conn, C_DISCONNECTING), 0);
+
+ switch (rv) {
+ case SS_NOTHING_TO_DO:
+ case SS_ALREADY_STANDALONE:
+ return SS_SUCCESS;
+ case SS_PRIMARY_NOP:
+ /* Our state checking code wants to see the peer outdated. */
+ rv = conn_request_state(tconn, NS2(conn, C_DISCONNECTING,
+ pdsk, D_OUTDATED), CS_VERBOSE);
+ break;
+ case SS_CW_FAILED_BY_PEER:
+ /* The peer probably wants to see us outdated. */
+ rv = conn_request_state(tconn, NS2(conn, C_DISCONNECTING,
+ disk, D_OUTDATED), 0);
+ if (rv == SS_IS_DISKLESS || rv == SS_LOWER_THAN_OUTDATED) {
+ conn_request_state(tconn, NS(conn, C_DISCONNECTING), CS_HARD);
+ rv = SS_SUCCESS;
+ }
+ break;
+ default:;
+ /* no special handling necessary */
}
- if (dc.force) {
- spin_lock_irq(&mdev->tconn->req_lock);
- if (mdev->state.conn >= C_WF_CONNECTION)
- _drbd_set_state(_NS(mdev, conn, C_DISCONNECTING), CS_HARD, NULL);
- spin_unlock_irq(&mdev->tconn->req_lock);
- goto done;
- }
+ return rv;
+}
- retcode = _drbd_request_state(mdev, NS(conn, C_DISCONNECTING), CS_ORDERED);
+int drbd_adm_disconnect(struct sk_buff *skb, struct genl_info *info)
+{
+ struct disconnect_parms parms;
+ struct drbd_tconn *tconn;
+ enum drbd_state_rv rv;
+ enum drbd_ret_code retcode;
+ int err;
- if (retcode == SS_NOTHING_TO_DO)
- goto done;
- else if (retcode == SS_ALREADY_STANDALONE)
- goto done;
- else if (retcode == SS_PRIMARY_NOP) {
- /* Our statche checking code wants to see the peer outdated. */
- retcode = drbd_request_state(mdev, NS2(conn, C_DISCONNECTING,
- pdsk, D_OUTDATED));
- } else if (retcode == SS_CW_FAILED_BY_PEER) {
- /* The peer probably wants to see us outdated. */
- retcode = _drbd_request_state(mdev, NS2(conn, C_DISCONNECTING,
- disk, D_OUTDATED),
- CS_ORDERED);
- if (retcode == SS_IS_DISKLESS || retcode == SS_LOWER_THAN_OUTDATED) {
- drbd_force_state(mdev, NS(conn, C_DISCONNECTING));
- retcode = SS_SUCCESS;
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto fail;
+
+ tconn = adm_ctx.tconn;
+ memset(&parms, 0, sizeof(parms));
+ if (info->attrs[DRBD_NLA_DISCONNECT_PARMS]) {
+ err = disconnect_parms_from_attrs(&parms, info);
+ if (err) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto fail;
}
}
- if (retcode < SS_SUCCESS)
+ rv = conn_try_disconnect(tconn, parms.force_disconnect);
+ if (rv < SS_SUCCESS)
goto fail;
- if (wait_event_interruptible(mdev->state_wait,
- mdev->state.conn != C_DISCONNECTING)) {
+ if (wait_event_interruptible(tconn->ping_wait,
+ tconn->cstate != C_DISCONNECTING)) {
/* Do not test for mdev->state.conn == C_STANDALONE, since
someone else might connect us in the mean time! */
retcode = ERR_INTR;
goto fail;
}
- done:
retcode = NO_ERROR;
fail:
- drbd_md_sync(mdev);
- reply->ret_code = retcode;
+ drbd_adm_finish(info, retcode);
return 0;
}
_drbd_request_state(mdev, NS(conn, C_WF_SYNC_UUID), CS_VERBOSE + CS_SERIALIZE);
}
-static int drbd_nl_resize(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_resize(struct sk_buff *skb, struct genl_info *info)
{
- struct resize rs;
- int retcode = NO_ERROR;
+ struct resize_parms rs;
+ struct drbd_conf *mdev;
+ enum drbd_ret_code retcode;
enum determine_dev_size dd;
enum dds_flags ddsf;
+ int err;
- memset(&rs, 0, sizeof(struct resize));
- if (!resize_from_tags(mdev, nlp->tag_list, &rs)) {
- retcode = ERR_MANDATORY_TAG;
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
goto fail;
+
+ memset(&rs, 0, sizeof(struct resize_parms));
+ if (info->attrs[DRBD_NLA_RESIZE_PARMS]) {
+ err = resize_parms_from_attrs(&rs, info);
+ if (err) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto fail;
+ }
}
+ mdev = adm_ctx.mdev;
if (mdev->state.conn > C_CONNECTED) {
retcode = ERR_RESIZE_RESYNC;
goto fail;
}
fail:
- reply->ret_code = retcode;
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_syncer_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_resource_opts(struct sk_buff *skb, struct genl_info *info)
{
- int retcode = NO_ERROR;
- int err;
- int ovr; /* online verify running */
- int rsr; /* re-sync running */
- struct crypto_hash *verify_tfm = NULL;
- struct crypto_hash *csums_tfm = NULL;
- struct syncer_conf sc;
+ enum drbd_ret_code retcode;
cpumask_var_t new_cpu_mask;
+ struct drbd_tconn *tconn;
int *rs_plan_s = NULL;
- int fifo_size;
+ struct res_opts sc;
+ int err;
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto fail;
+ tconn = adm_ctx.tconn;
if (!zalloc_cpumask_var(&new_cpu_mask, GFP_KERNEL)) {
retcode = ERR_NOMEM;
+ drbd_msg_put_info("unable to allocate cpumask");
goto fail;
}
- if (nlp->flags & DRBD_NL_SET_DEFAULTS) {
- memset(&sc, 0, sizeof(struct syncer_conf));
- sc.rate = DRBD_RATE_DEF;
- sc.after = DRBD_AFTER_DEF;
- sc.al_extents = DRBD_AL_EXTENTS_DEF;
+ if (((struct drbd_genlmsghdr*)info->userhdr)->flags
+ & DRBD_GENL_F_SET_DEFAULTS) {
+ memset(&sc, 0, sizeof(struct res_opts));
sc.on_no_data = DRBD_ON_NO_DATA_DEF;
- sc.c_plan_ahead = DRBD_C_PLAN_AHEAD_DEF;
- sc.c_delay_target = DRBD_C_DELAY_TARGET_DEF;
- sc.c_fill_target = DRBD_C_FILL_TARGET_DEF;
- sc.c_max_rate = DRBD_C_MAX_RATE_DEF;
- sc.c_min_rate = DRBD_C_MIN_RATE_DEF;
} else
- memcpy(&sc, &mdev->sync_conf, sizeof(struct syncer_conf));
+ sc = tconn->res_opts;
- if (!syncer_conf_from_tags(mdev, nlp->tag_list, &sc)) {
+ err = res_opts_from_attrs(&sc, info);
+ if (err) {
retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
goto fail;
}
- /* re-sync running */
- rsr = ( mdev->state.conn == C_SYNC_SOURCE ||
- mdev->state.conn == C_SYNC_TARGET ||
- mdev->state.conn == C_PAUSED_SYNC_S ||
- mdev->state.conn == C_PAUSED_SYNC_T );
-
- if (rsr && strcmp(sc.csums_alg, mdev->sync_conf.csums_alg)) {
- retcode = ERR_CSUMS_RESYNC_RUNNING;
- goto fail;
- }
-
- if (!rsr && sc.csums_alg[0]) {
- csums_tfm = crypto_alloc_hash(sc.csums_alg, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(csums_tfm)) {
- csums_tfm = NULL;
- retcode = ERR_CSUMS_ALG;
- goto fail;
- }
-
- if (!drbd_crypto_is_hash(crypto_hash_tfm(csums_tfm))) {
- retcode = ERR_CSUMS_ALG_ND;
- goto fail;
- }
- }
-
- /* online verify running */
- ovr = (mdev->state.conn == C_VERIFY_S || mdev->state.conn == C_VERIFY_T);
-
- if (ovr) {
- if (strcmp(sc.verify_alg, mdev->sync_conf.verify_alg)) {
- retcode = ERR_VERIFY_RUNNING;
- goto fail;
- }
- }
-
- if (!ovr && sc.verify_alg[0]) {
- verify_tfm = crypto_alloc_hash(sc.verify_alg, 0, CRYPTO_ALG_ASYNC);
- if (IS_ERR(verify_tfm)) {
- verify_tfm = NULL;
- retcode = ERR_VERIFY_ALG;
- goto fail;
- }
-
- if (!drbd_crypto_is_hash(crypto_hash_tfm(verify_tfm))) {
- retcode = ERR_VERIFY_ALG_ND;
- goto fail;
- }
- }
-
- /* silently ignore cpu mask on UP kernel */
- if (nr_cpu_ids > 1 && sc.cpu_mask[0] != 0) {
- err = __bitmap_parse(sc.cpu_mask, 32, 0,
- cpumask_bits(new_cpu_mask), nr_cpu_ids);
- if (err) {
- dev_warn(DEV, "__bitmap_parse() failed with %d\n", err);
- retcode = ERR_CPU_MASK_PARSE;
- goto fail;
- }
- }
-
- if (!expect(sc.rate >= 1))
- sc.rate = 1;
- if (!expect(sc.al_extents >= 7))
- sc.al_extents = 127; /* arbitrary minimum */
-#define AL_MAX ((MD_AL_MAX_SIZE-1) * AL_EXTENTS_PT)
- if (sc.al_extents > AL_MAX) {
- dev_err(DEV, "sc.al_extents > %d\n", AL_MAX);
- sc.al_extents = AL_MAX;
- }
-#undef AL_MAX
-
- /* to avoid spurious errors when configuring minors before configuring
- * the minors they depend on: if necessary, first create the minor we
- * depend on */
- if (sc.after >= 0)
- ensure_mdev(sc.after, 1);
-
- /* most sanity checks done, try to assign the new sync-after
- * dependency. need to hold the global lock in there,
- * to avoid a race in the dependency loop check. */
- retcode = drbd_alter_sa(mdev, sc.after);
- if (retcode != NO_ERROR)
- goto fail;
-
- fifo_size = (sc.c_plan_ahead * 10 * SLEEP_TIME) / HZ;
- if (fifo_size != mdev->rs_plan_s.size && fifo_size > 0) {
- rs_plan_s = kzalloc(sizeof(int) * fifo_size, GFP_KERNEL);
- if (!rs_plan_s) {
- dev_err(DEV, "kmalloc of fifo_buffer failed");
- retcode = ERR_NOMEM;
- goto fail;
- }
- }
-
- /* ok, assign the rest of it as well.
- * lock against receive_SyncParam() */
- spin_lock(&mdev->peer_seq_lock);
- mdev->sync_conf = sc;
-
- if (!rsr) {
- crypto_free_hash(mdev->csums_tfm);
- mdev->csums_tfm = csums_tfm;
- csums_tfm = NULL;
- }
-
- if (!ovr) {
- crypto_free_hash(mdev->verify_tfm);
- mdev->verify_tfm = verify_tfm;
- verify_tfm = NULL;
- }
-
- if (fifo_size != mdev->rs_plan_s.size) {
- kfree(mdev->rs_plan_s.values);
- mdev->rs_plan_s.values = rs_plan_s;
- mdev->rs_plan_s.size = fifo_size;
- mdev->rs_planed = 0;
- rs_plan_s = NULL;
- }
-
- spin_unlock(&mdev->peer_seq_lock);
-
- if (get_ldev(mdev)) {
- wait_event(mdev->al_wait, lc_try_lock(mdev->act_log));
- drbd_al_shrink(mdev);
- err = drbd_check_al_size(mdev);
- lc_unlock(mdev->act_log);
- wake_up(&mdev->al_wait);
-
- put_ldev(mdev);
- drbd_md_sync(mdev);
-
+ /* silently ignore cpu mask on UP kernel */
+ if (nr_cpu_ids > 1 && sc.cpu_mask[0] != 0) {
+ err = __bitmap_parse(sc.cpu_mask, 32, 0,
+ cpumask_bits(new_cpu_mask), nr_cpu_ids);
if (err) {
- retcode = ERR_NOMEM;
+ conn_warn(tconn, "__bitmap_parse() failed with %d\n", err);
+ retcode = ERR_CPU_MASK_PARSE;
goto fail;
}
}
- if (mdev->state.conn >= C_CONNECTED)
- drbd_send_sync_param(mdev, &sc);
- if (!cpumask_equal(mdev->cpu_mask, new_cpu_mask)) {
- cpumask_copy(mdev->cpu_mask, new_cpu_mask);
- drbd_calc_cpu_mask(mdev);
- mdev->tconn->receiver.reset_cpu_mask = 1;
- mdev->tconn->asender.reset_cpu_mask = 1;
- mdev->tconn->worker.reset_cpu_mask = 1;
+ tconn->res_opts = sc;
+
+ if (!cpumask_equal(tconn->cpu_mask, new_cpu_mask)) {
+ cpumask_copy(tconn->cpu_mask, new_cpu_mask);
+ drbd_calc_cpu_mask(tconn);
+ tconn->receiver.reset_cpu_mask = 1;
+ tconn->asender.reset_cpu_mask = 1;
+ tconn->worker.reset_cpu_mask = 1;
}
- kobject_uevent(&disk_to_dev(mdev->vdisk)->kobj, KOBJ_CHANGE);
fail:
kfree(rs_plan_s);
free_cpumask_var(new_cpu_mask);
- crypto_free_hash(csums_tfm);
- crypto_free_hash(verify_tfm);
- reply->ret_code = retcode;
+
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_invalidate(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_invalidate(struct sk_buff *skb, struct genl_info *info)
{
- int retcode;
+ struct drbd_conf *mdev;
+ int retcode; /* enum drbd_ret_code rsp. enum drbd_state_rv */
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ mdev = adm_ctx.mdev;
/* If there is still bitmap IO pending, probably because of a previous
* resync just being finished, wait for it before requesting a new resync. */
retcode = drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_T));
}
- reply->ret_code = retcode;
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
return rv;
}
-static int drbd_nl_invalidate_peer(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+static int drbd_adm_simple_request_state(struct sk_buff *skb, struct genl_info *info,
+ union drbd_state mask, union drbd_state val)
{
- int retcode;
-
- /* If there is still bitmap IO pending, probably because of a previous
- * resync just being finished, wait for it before requesting a new resync. */
- wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags));
+ enum drbd_ret_code retcode;
- retcode = _drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_S), CS_ORDERED);
-
- if (retcode < SS_SUCCESS) {
- if (retcode == SS_NEED_CONNECTION && mdev->state.role == R_PRIMARY) {
- /* The peer will get a resync upon connect anyways. Just make that
- into a full resync. */
- retcode = drbd_request_state(mdev, NS(pdsk, D_INCONSISTENT));
- if (retcode >= SS_SUCCESS) {
- if (drbd_bitmap_io(mdev, &drbd_bmio_set_susp_al,
- "set_n_write from invalidate_peer",
- BM_LOCKED_SET_ALLOWED))
- retcode = ERR_IO_MD_DISK;
- }
- } else
- retcode = drbd_request_state(mdev, NS(conn, C_STARTING_SYNC_S));
- }
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
- reply->ret_code = retcode;
+ retcode = drbd_request_state(adm_ctx.mdev, mask, val);
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_pause_sync(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_invalidate_peer(struct sk_buff *skb, struct genl_info *info)
{
- int retcode = NO_ERROR;
+ return drbd_adm_simple_request_state(skb, info, NS(conn, C_STARTING_SYNC_S));
+}
- if (drbd_request_state(mdev, NS(user_isp, 1)) == SS_NOTHING_TO_DO)
- retcode = ERR_PAUSE_IS_SET;
+int drbd_adm_pause_sync(struct sk_buff *skb, struct genl_info *info)
+{
+ enum drbd_ret_code retcode;
- reply->ret_code = retcode;
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ if (drbd_request_state(adm_ctx.mdev, NS(user_isp, 1)) == SS_NOTHING_TO_DO)
+ retcode = ERR_PAUSE_IS_SET;
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_resume_sync(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_resume_sync(struct sk_buff *skb, struct genl_info *info)
{
- int retcode = NO_ERROR;
- union drbd_state s;
+ union drbd_dev_state s;
+ enum drbd_ret_code retcode;
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
- if (drbd_request_state(mdev, NS(user_isp, 0)) == SS_NOTHING_TO_DO) {
- s = mdev->state;
+ if (drbd_request_state(adm_ctx.mdev, NS(user_isp, 0)) == SS_NOTHING_TO_DO) {
+ s = adm_ctx.mdev->state;
if (s.conn == C_PAUSED_SYNC_S || s.conn == C_PAUSED_SYNC_T) {
retcode = s.aftr_isp ? ERR_PIC_AFTER_DEP :
s.peer_isp ? ERR_PIC_PEER_DEP : ERR_PAUSE_IS_CLEAR;
}
}
- reply->ret_code = retcode;
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_suspend_io(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_suspend_io(struct sk_buff *skb, struct genl_info *info)
{
- reply->ret_code = drbd_request_state(mdev, NS(susp, 1));
-
- return 0;
+ return drbd_adm_simple_request_state(skb, info, NS(susp, 1));
}
-static int drbd_nl_resume_io(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_resume_io(struct sk_buff *skb, struct genl_info *info)
{
+ struct drbd_conf *mdev;
+ int retcode; /* enum drbd_ret_code rsp. enum drbd_state_rv */
+
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+
+ mdev = adm_ctx.mdev;
if (test_bit(NEW_CUR_UUID, &mdev->flags)) {
drbd_uuid_new_current(mdev);
clear_bit(NEW_CUR_UUID, &mdev->flags);
}
drbd_suspend_io(mdev);
- reply->ret_code = drbd_request_state(mdev, NS3(susp, 0, susp_nod, 0, susp_fen, 0));
- if (reply->ret_code == SS_SUCCESS) {
+ retcode = drbd_request_state(mdev, NS3(susp, 0, susp_nod, 0, susp_fen, 0));
+ if (retcode == SS_SUCCESS) {
if (mdev->state.conn < C_CONNECTED)
- tl_clear(mdev);
+ tl_clear(mdev->tconn);
if (mdev->state.disk == D_DISKLESS || mdev->state.disk == D_FAILED)
- tl_restart(mdev, FAIL_FROZEN_DISK_IO);
+ tl_restart(mdev->tconn, FAIL_FROZEN_DISK_IO);
}
drbd_resume_io(mdev);
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_outdate(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_outdate(struct sk_buff *skb, struct genl_info *info)
{
- reply->ret_code = drbd_request_state(mdev, NS(disk, D_OUTDATED));
- return 0;
+ return drbd_adm_simple_request_state(skb, info, NS(disk, D_OUTDATED));
}
-static int drbd_nl_get_config(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int nla_put_drbd_cfg_context(struct sk_buff *skb, const char *conn_name, unsigned vnr)
{
- unsigned short *tl;
+ struct nlattr *nla;
+ nla = nla_nest_start(skb, DRBD_NLA_CFG_CONTEXT);
+ if (!nla)
+ goto nla_put_failure;
+ if (vnr != VOLUME_UNSPECIFIED)
+ NLA_PUT_U32(skb, T_ctx_volume, vnr);
+ NLA_PUT_STRING(skb, T_ctx_conn_name, conn_name);
+ nla_nest_end(skb, nla);
+ return 0;
- tl = reply->tag_list;
+nla_put_failure:
+ if (nla)
+ nla_nest_cancel(skb, nla);
+ return -EMSGSIZE;
+}
- if (get_ldev(mdev)) {
- tl = disk_conf_to_tags(mdev, &mdev->ldev->dc, tl);
- put_ldev(mdev);
+int nla_put_status_info(struct sk_buff *skb, struct drbd_conf *mdev,
+ const struct sib_info *sib)
+{
+ struct state_info *si = NULL; /* for sizeof(si->member); */
+ struct net_conf *nc;
+ struct nlattr *nla;
+ int got_ldev;
+ int err = 0;
+ int exclude_sensitive;
+
+ /* If sib != NULL, this is drbd_bcast_event, which anyone can listen
+ * to. So we better exclude_sensitive information.
+ *
+ * If sib == NULL, this is drbd_adm_get_status, executed synchronously
+ * in the context of the requesting user process. Exclude sensitive
+ * information, unless current has superuser.
+ *
+ * NOTE: for drbd_adm_get_status_all(), this is a netlink dump, and
+ * relies on the current implementation of netlink_dump(), which
+ * executes the dump callback successively from netlink_recvmsg(),
+ * always in the context of the receiving process */
+ exclude_sensitive = sib || !capable(CAP_SYS_ADMIN);
+
+ got_ldev = get_ldev(mdev);
+
+ /* We need to add connection name and volume number information still.
+ * Minor number is in drbd_genlmsghdr. */
+ if (nla_put_drbd_cfg_context(skb, mdev->tconn->name, mdev->vnr))
+ goto nla_put_failure;
+
+ if (res_opts_to_skb(skb, &mdev->tconn->res_opts, exclude_sensitive))
+ goto nla_put_failure;
+
+ if (got_ldev)
+ if (disk_conf_to_skb(skb, &mdev->ldev->dc, exclude_sensitive))
+ goto nla_put_failure;
+
+ rcu_read_lock();
+ nc = rcu_dereference(mdev->tconn->net_conf);
+ if (nc)
+ err = net_conf_to_skb(skb, nc, exclude_sensitive);
+ rcu_read_unlock();
+ if (err)
+ goto nla_put_failure;
+
+ nla = nla_nest_start(skb, DRBD_NLA_STATE_INFO);
+ if (!nla)
+ goto nla_put_failure;
+ NLA_PUT_U32(skb, T_sib_reason, sib ? sib->sib_reason : SIB_GET_STATUS_REPLY);
+ NLA_PUT_U32(skb, T_current_state, mdev->state.i);
+ NLA_PUT_U64(skb, T_ed_uuid, mdev->ed_uuid);
+ NLA_PUT_U64(skb, T_capacity, drbd_get_capacity(mdev->this_bdev));
+
+ if (got_ldev) {
+ NLA_PUT_U32(skb, T_disk_flags, mdev->ldev->md.flags);
+ NLA_PUT(skb, T_uuids, sizeof(si->uuids), mdev->ldev->md.uuid);
+ NLA_PUT_U64(skb, T_bits_total, drbd_bm_bits(mdev));
+ NLA_PUT_U64(skb, T_bits_oos, drbd_bm_total_weight(mdev));
+ if (C_SYNC_SOURCE <= mdev->state.conn &&
+ C_PAUSED_SYNC_T >= mdev->state.conn) {
+ NLA_PUT_U64(skb, T_bits_rs_total, mdev->rs_total);
+ NLA_PUT_U64(skb, T_bits_rs_failed, mdev->rs_failed);
+ }
}
- if (get_net_conf(mdev->tconn)) {
- tl = net_conf_to_tags(mdev, mdev->tconn->net_conf, tl);
- put_net_conf(mdev->tconn);
+ if (sib) {
+ switch(sib->sib_reason) {
+ case SIB_SYNC_PROGRESS:
+ case SIB_GET_STATUS_REPLY:
+ break;
+ case SIB_STATE_CHANGE:
+ NLA_PUT_U32(skb, T_prev_state, sib->os.i);
+ NLA_PUT_U32(skb, T_new_state, sib->ns.i);
+ break;
+ case SIB_HELPER_POST:
+ NLA_PUT_U32(skb,
+ T_helper_exit_code, sib->helper_exit_code);
+ /* fall through */
+ case SIB_HELPER_PRE:
+ NLA_PUT_STRING(skb, T_helper, sib->helper_name);
+ break;
+ }
}
- tl = syncer_conf_to_tags(mdev, &mdev->sync_conf, tl);
+ nla_nest_end(skb, nla);
- put_unaligned(TT_END, tl++); /* Close the tag list */
-
- return (int)((char *)tl - (char *)reply->tag_list);
+ if (0)
+nla_put_failure:
+ err = -EMSGSIZE;
+ if (got_ldev)
+ put_ldev(mdev);
+ return err;
}
-static int drbd_nl_get_state(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_get_status(struct sk_buff *skb, struct genl_info *info)
{
- unsigned short *tl = reply->tag_list;
- union drbd_state s = mdev->state;
- unsigned long rs_left;
- unsigned int res;
+ enum drbd_ret_code retcode;
+ int err;
- tl = get_state_to_tags(mdev, (struct get_state *)&s, tl);
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
- /* no local ref, no bitmap, no syncer progress. */
- if (s.conn >= C_SYNC_SOURCE && s.conn <= C_PAUSED_SYNC_T) {
- if (get_ldev(mdev)) {
- drbd_get_syncer_progress(mdev, &rs_left, &res);
- tl = tl_add_int(tl, T_sync_progress, &res);
- put_ldev(mdev);
- }
+ err = nla_put_status_info(adm_ctx.reply_skb, adm_ctx.mdev, NULL);
+ if (err) {
+ nlmsg_free(adm_ctx.reply_skb);
+ return err;
}
- put_unaligned(TT_END, tl++); /* Close the tag list */
-
- return (int)((char *)tl - (char *)reply->tag_list);
+out:
+ drbd_adm_finish(info, retcode);
+ return 0;
}
-static int drbd_nl_get_uuids(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int get_one_status(struct sk_buff *skb, struct netlink_callback *cb)
{
- unsigned short *tl;
-
- tl = reply->tag_list;
+ struct drbd_conf *mdev;
+ struct drbd_genlmsghdr *dh;
+ struct drbd_tconn *pos = (struct drbd_tconn*)cb->args[0];
+ struct drbd_tconn *tconn = NULL;
+ struct drbd_tconn *tmp;
+ unsigned volume = cb->args[1];
+
+ /* Open coded, deferred, iteration:
+ * list_for_each_entry_safe(tconn, tmp, &drbd_tconns, all_tconn) {
+ * idr_for_each_entry(&tconn->volumes, mdev, i) {
+ * ...
+ * }
+ * }
+ * where tconn is cb->args[0];
+ * and i is cb->args[1];
+ *
+ * cb->args[2] indicates if we shall loop over all resources,
+ * or just dump all volumes of a single resource.
+ *
+ * This may miss entries inserted after this dump started,
+ * or entries deleted before they are reached.
+ *
+ * We need to make sure the mdev won't disappear while
+ * we are looking at it, and revalidate our iterators
+ * on each iteration.
+ */
- if (get_ldev(mdev)) {
- tl = tl_add_blob(tl, T_uuids, mdev->ldev->md.uuid, UI_SIZE*sizeof(u64));
- tl = tl_add_int(tl, T_uuids_flags, &mdev->ldev->md.flags);
- put_ldev(mdev);
+ /* synchronize with conn_create()/conn_destroy() */
+ down_read(&drbd_cfg_rwsem);
+ /* revalidate iterator position */
+ list_for_each_entry(tmp, &drbd_tconns, all_tconn) {
+ if (pos == NULL) {
+ /* first iteration */
+ pos = tmp;
+ tconn = pos;
+ break;
+ }
+ if (tmp == pos) {
+ tconn = pos;
+ break;
+ }
}
- put_unaligned(TT_END, tl++); /* Close the tag list */
+ if (tconn) {
+next_tconn:
+ mdev = idr_get_next(&tconn->volumes, &volume);
+ if (!mdev) {
+ /* No more volumes to dump on this tconn.
+ * Advance tconn iterator. */
+ pos = list_entry(tconn->all_tconn.next,
+ struct drbd_tconn, all_tconn);
+ /* Did we dump any volume on this tconn yet? */
+ if (volume != 0) {
+ /* If we reached the end of the list,
+ * or only a single resource dump was requested,
+ * we are done. */
+ if (&pos->all_tconn == &drbd_tconns || cb->args[2])
+ goto out;
+ volume = 0;
+ tconn = pos;
+ goto next_tconn;
+ }
+ }
+
+ dh = genlmsg_put(skb, NETLINK_CB(cb->skb).pid,
+ cb->nlh->nlmsg_seq, &drbd_genl_family,
+ NLM_F_MULTI, DRBD_ADM_GET_STATUS);
+ if (!dh)
+ goto out;
+
+ if (!mdev) {
+ /* this is a tconn without a single volume */
+ dh->minor = -1U;
+ dh->ret_code = NO_ERROR;
+ if (nla_put_drbd_cfg_context(skb, tconn->name, VOLUME_UNSPECIFIED))
+ genlmsg_cancel(skb, dh);
+ else
+ genlmsg_end(skb, dh);
+ goto out;
+ }
+
+ D_ASSERT(mdev->vnr == volume);
+ D_ASSERT(mdev->tconn == tconn);
+
+ dh->minor = mdev_to_minor(mdev);
+ dh->ret_code = NO_ERROR;
- return (int)((char *)tl - (char *)reply->tag_list);
+ if (nla_put_status_info(skb, mdev, NULL)) {
+ genlmsg_cancel(skb, dh);
+ goto out;
+ }
+ genlmsg_end(skb, dh);
+ }
+
+out:
+ up_read(&drbd_cfg_rwsem);
+ /* where to start the next iteration */
+ cb->args[0] = (long)pos;
+ cb->args[1] = (pos == tconn) ? volume + 1 : 0;
+
+ /* No more tconns/volumes/minors found results in an empty skb.
+ * Which will terminate the dump. */
+ return skb->len;
}
-/**
- * drbd_nl_get_timeout_flag() - Used by drbdsetup to find out which timeout value to use
- * @mdev: DRBD device.
- * @nlp: Netlink/connector packet from drbdsetup
- * @reply: Reply packet for drbdsetup
+/*
+ * Request status of all resources, or of all volumes within a single resource.
+ *
+ * This is a dump, as the answer may not fit in a single reply skb otherwise.
+ * Which means we cannot use the family->attrbuf or other such members, because
+ * dump is NOT protected by the genl_lock(). During dump, we only have access
+ * to the incoming skb, and need to opencode "parsing" of the nlattr payload.
+ *
+ * Once things are setup properly, we call into get_one_status().
*/
-static int drbd_nl_get_timeout_flag(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_get_status_all(struct sk_buff *skb, struct netlink_callback *cb)
{
- unsigned short *tl;
- char rv;
+ const unsigned hdrlen = GENL_HDRLEN + GENL_MAGIC_FAMILY_HDRSZ;
+ struct nlattr *nla;
+ const char *conn_name;
+ struct drbd_tconn *tconn;
+
+ /* Is this a followup call? */
+ if (cb->args[0]) {
+ /* ... of a single resource dump,
+ * and the resource iterator has been advanced already? */
+ if (cb->args[2] && cb->args[2] != cb->args[0])
+ return 0; /* DONE. */
+ goto dump;
+ }
+
+ /* First call (from netlink_dump_start). We need to figure out
+ * which resource(s) the user wants us to dump. */
+ nla = nla_find(nlmsg_attrdata(cb->nlh, hdrlen),
+ nlmsg_attrlen(cb->nlh, hdrlen),
+ DRBD_NLA_CFG_CONTEXT);
+
+ /* No explicit context given. Dump all. */
+ if (!nla)
+ goto dump;
+ nla = nla_find_nested(nla, __nla_type(T_ctx_conn_name));
+ /* context given, but no name present? */
+ if (!nla)
+ return -EINVAL;
+ conn_name = nla_data(nla);
+ tconn = conn_by_name(conn_name);
+ if (!tconn)
+ return -ENODEV;
+
+ /* prime iterators, and set "filter" mode mark:
+ * only dump this tconn. */
+ cb->args[0] = (long)tconn;
+ /* cb->args[1] = 0; passed in this way. */
+ cb->args[2] = (long)tconn;
+
+dump:
+ return get_one_status(skb, cb);
+}
- tl = reply->tag_list;
+int drbd_adm_get_timeout_type(struct sk_buff *skb, struct genl_info *info)
+{
+ enum drbd_ret_code retcode;
+ struct timeout_parms tp;
+ int err;
- rv = mdev->state.pdsk == D_OUTDATED ? UT_PEER_OUTDATED :
- test_bit(USE_DEGR_WFC_T, &mdev->flags) ? UT_DEGRADED : UT_DEFAULT;
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
- tl = tl_add_blob(tl, T_use_degraded, &rv, sizeof(rv));
- put_unaligned(TT_END, tl++); /* Close the tag list */
+ tp.timeout_type =
+ adm_ctx.mdev->state.pdsk == D_OUTDATED ? UT_PEER_OUTDATED :
+ test_bit(USE_DEGR_WFC_T, &adm_ctx.mdev->flags) ? UT_DEGRADED :
+ UT_DEFAULT;
- return (int)((char *)tl - (char *)reply->tag_list);
+ err = timeout_parms_to_priv_skb(adm_ctx.reply_skb, &tp);
+ if (err) {
+ nlmsg_free(adm_ctx.reply_skb);
+ return err;
+ }
+out:
+ drbd_adm_finish(info, retcode);
+ return 0;
}
-static int drbd_nl_start_ov(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_start_ov(struct sk_buff *skb, struct genl_info *info)
{
- /* default to resume from last known position, if possible */
- struct start_ov args =
- { .start_sector = mdev->ov_start_sector };
+ struct drbd_conf *mdev;
+ enum drbd_ret_code retcode;
- if (!start_ov_from_tags(mdev, nlp->tag_list, &args)) {
- reply->ret_code = ERR_MANDATORY_TAG;
- return 0;
- }
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
+ mdev = adm_ctx.mdev;
+ if (info->attrs[DRBD_NLA_START_OV_PARMS]) {
+ /* resume from last known position, if possible */
+ struct start_ov_parms parms =
+ { .ov_start_sector = mdev->ov_start_sector };
+ int err = start_ov_parms_from_attrs(&parms, info);
+ if (err) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto out;
+ }
+ /* w_make_ov_request expects position to be aligned */
+ mdev->ov_start_sector = parms.ov_start_sector & ~BM_SECT_PER_BIT;
+ }
/* If there is still bitmap IO pending, e.g. previous resync or verify
* just being finished, wait for it before requesting a new resync. */
wait_event(mdev->misc_wait, !test_bit(BITMAP_IO, &mdev->flags));
-
- /* w_make_ov_request expects position to be aligned */
- mdev->ov_start_sector = args.start_sector & ~BM_SECT_PER_BIT;
- reply->ret_code = drbd_request_state(mdev,NS(conn,C_VERIFY_S));
+ retcode = drbd_request_state(mdev,NS(conn,C_VERIFY_S));
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
-static int drbd_nl_new_c_uuid(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
- struct drbd_nl_cfg_reply *reply)
+int drbd_adm_new_c_uuid(struct sk_buff *skb, struct genl_info *info)
{
- int retcode = NO_ERROR;
+ struct drbd_conf *mdev;
+ enum drbd_ret_code retcode;
int skip_initial_sync = 0;
int err;
+ struct new_c_uuid_parms args;
- struct new_c_uuid args;
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out_nolock;
- memset(&args, 0, sizeof(struct new_c_uuid));
- if (!new_c_uuid_from_tags(mdev, nlp->tag_list, &args)) {
- reply->ret_code = ERR_MANDATORY_TAG;
- return 0;
+ mdev = adm_ctx.mdev;
+ memset(&args, 0, sizeof(args));
+ if (info->attrs[DRBD_NLA_NEW_C_UUID_PARMS]) {
+ err = new_c_uuid_parms_from_attrs(&args, info);
+ if (err) {
+ retcode = ERR_MANDATORY_TAG;
+ drbd_msg_put_info(from_attrs_err_to_txt(err));
+ goto out_nolock;
+ }
}
- mutex_lock(&mdev->state_mutex); /* Protects us against serialized state changes. */
+ mutex_lock(mdev->state_mutex); /* Protects us against serialized state changes. */
if (!get_ldev(mdev)) {
retcode = ERR_NO_DISK;
out_dec:
put_ldev(mdev);
out:
- mutex_unlock(&mdev->state_mutex);
-
- reply->ret_code = retcode;
+ mutex_unlock(mdev->state_mutex);
+out_nolock:
+ drbd_adm_finish(info, retcode);
return 0;
}
-struct cn_handler_struct {
- int (*function)(struct drbd_conf *,
- struct drbd_nl_cfg_req *,
- struct drbd_nl_cfg_reply *);
- int reply_body_size;
-};
-
-static struct cn_handler_struct cnd_table[] = {
- [ P_primary ] = { &drbd_nl_primary, 0 },
- [ P_secondary ] = { &drbd_nl_secondary, 0 },
- [ P_disk_conf ] = { &drbd_nl_disk_conf, 0 },
- [ P_detach ] = { &drbd_nl_detach, 0 },
- [ P_net_conf ] = { &drbd_nl_net_conf, 0 },
- [ P_disconnect ] = { &drbd_nl_disconnect, 0 },
- [ P_resize ] = { &drbd_nl_resize, 0 },
- [ P_syncer_conf ] = { &drbd_nl_syncer_conf, 0 },
- [ P_invalidate ] = { &drbd_nl_invalidate, 0 },
- [ P_invalidate_peer ] = { &drbd_nl_invalidate_peer, 0 },
- [ P_pause_sync ] = { &drbd_nl_pause_sync, 0 },
- [ P_resume_sync ] = { &drbd_nl_resume_sync, 0 },
- [ P_suspend_io ] = { &drbd_nl_suspend_io, 0 },
- [ P_resume_io ] = { &drbd_nl_resume_io, 0 },
- [ P_outdate ] = { &drbd_nl_outdate, 0 },
- [ P_get_config ] = { &drbd_nl_get_config,
- sizeof(struct syncer_conf_tag_len_struct) +
- sizeof(struct disk_conf_tag_len_struct) +
- sizeof(struct net_conf_tag_len_struct) },
- [ P_get_state ] = { &drbd_nl_get_state,
- sizeof(struct get_state_tag_len_struct) +
- sizeof(struct sync_progress_tag_len_struct) },
- [ P_get_uuids ] = { &drbd_nl_get_uuids,
- sizeof(struct get_uuids_tag_len_struct) },
- [ P_get_timeout_flag ] = { &drbd_nl_get_timeout_flag,
- sizeof(struct get_timeout_flag_tag_len_struct)},
- [ P_start_ov ] = { &drbd_nl_start_ov, 0 },
- [ P_new_c_uuid ] = { &drbd_nl_new_c_uuid, 0 },
-};
-
-static void drbd_connector_callback(struct cn_msg *req, struct netlink_skb_parms *nsp)
-{
- struct drbd_nl_cfg_req *nlp = (struct drbd_nl_cfg_req *)req->data;
- struct cn_handler_struct *cm;
- struct cn_msg *cn_reply;
- struct drbd_nl_cfg_reply *reply;
- struct drbd_conf *mdev;
- int retcode, rr;
- int reply_size = sizeof(struct cn_msg)
- + sizeof(struct drbd_nl_cfg_reply)
- + sizeof(short int);
-
- if (!try_module_get(THIS_MODULE)) {
- printk(KERN_ERR "drbd: try_module_get() failed!\n");
- return;
+static enum drbd_ret_code
+drbd_check_conn_name(const char *name)
+{
+ if (!name || !name[0]) {
+ drbd_msg_put_info("connection name missing");
+ return ERR_MANDATORY_TAG;
}
-
- if (!cap_raised(current_cap(), CAP_SYS_ADMIN)) {
- retcode = ERR_PERM;
- goto fail;
+ /* if we want to use these in sysfs/configfs/debugfs some day,
+ * we must not allow slashes */
+ if (strchr(name, '/')) {
+ drbd_msg_put_info("invalid connection name");
+ return ERR_INVALID_REQUEST;
}
+ return NO_ERROR;
+}
- mdev = ensure_mdev(nlp->drbd_minor,
- (nlp->flags & DRBD_NL_CREATE_DEVICE));
- if (!mdev) {
- retcode = ERR_MINOR_INVALID;
- goto fail;
- }
+int drbd_adm_create_connection(struct sk_buff *skb, struct genl_info *info)
+{
+ enum drbd_ret_code retcode;
- if (nlp->packet_type >= P_nl_after_last_packet ||
- nlp->packet_type == P_return_code_only) {
- retcode = ERR_PACKET_NR;
- goto fail;
- }
+ retcode = drbd_adm_prepare(skb, info, 0);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
- cm = cnd_table + nlp->packet_type;
+ retcode = drbd_check_conn_name(adm_ctx.conn_name);
+ if (retcode != NO_ERROR)
+ goto out;
- /* This may happen if packet number is 0: */
- if (cm->function == NULL) {
- retcode = ERR_PACKET_NR;
- goto fail;
+ if (adm_ctx.tconn) {
+ if (info->nlhdr->nlmsg_flags & NLM_F_EXCL) {
+ retcode = ERR_INVALID_REQUEST;
+ drbd_msg_put_info("connection exists");
+ }
+ /* else: still NO_ERROR */
+ goto out;
}
- reply_size += cm->reply_body_size;
-
- /* allocation not in the IO path, cqueue thread context */
- cn_reply = kzalloc(reply_size, GFP_KERNEL);
- if (!cn_reply) {
+ if (!conn_create(adm_ctx.conn_name))
retcode = ERR_NOMEM;
- goto fail;
- }
- reply = (struct drbd_nl_cfg_reply *) cn_reply->data;
-
- reply->packet_type =
- cm->reply_body_size ? nlp->packet_type : P_return_code_only;
- reply->minor = nlp->drbd_minor;
- reply->ret_code = NO_ERROR; /* Might by modified by cm->function. */
- /* reply->tag_list; might be modified by cm->function. */
-
- rr = cm->function(mdev, nlp, reply);
-
- cn_reply->id = req->id;
- cn_reply->seq = req->seq;
- cn_reply->ack = req->ack + 1;
- cn_reply->len = sizeof(struct drbd_nl_cfg_reply) + rr;
- cn_reply->flags = 0;
-
- rr = cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_KERNEL);
- if (rr && rr != -ESRCH)
- printk(KERN_INFO "drbd: cn_netlink_send()=%d\n", rr);
-
- kfree(cn_reply);
- module_put(THIS_MODULE);
- return;
- fail:
- drbd_nl_send_reply(req, retcode);
- module_put(THIS_MODULE);
+out:
+ drbd_adm_finish(info, retcode);
+ return 0;
}
-static atomic_t drbd_nl_seq = ATOMIC_INIT(2); /* two. */
-
-static unsigned short *
-__tl_add_blob(unsigned short *tl, enum drbd_tags tag, const void *data,
- unsigned short len, int nul_terminated)
+int drbd_adm_add_minor(struct sk_buff *skb, struct genl_info *info)
{
- unsigned short l = tag_descriptions[tag_number(tag)].max_len;
- len = (len < l) ? len : l;
- put_unaligned(tag, tl++);
- put_unaligned(len, tl++);
- memcpy(tl, data, len);
- tl = (unsigned short*)((char*)tl + len);
- if (nul_terminated)
- *((char*)tl - 1) = 0;
- return tl;
-}
+ struct drbd_genlmsghdr *dh = info->userhdr;
+ enum drbd_ret_code retcode;
-static unsigned short *
-tl_add_blob(unsigned short *tl, enum drbd_tags tag, const void *data, int len)
-{
- return __tl_add_blob(tl, tag, data, len, 0);
-}
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
-static unsigned short *
-tl_add_str(unsigned short *tl, enum drbd_tags tag, const char *str)
-{
- return __tl_add_blob(tl, tag, str, strlen(str)+1, 0);
-}
+ /* FIXME drop minor_count parameter, limit to MINORMASK */
+ if (dh->minor >= minor_count) {
+ drbd_msg_put_info("requested minor out of range");
+ retcode = ERR_INVALID_REQUEST;
+ goto out;
+ }
+ if (adm_ctx.volume > DRBD_VOLUME_MAX) {
+ drbd_msg_put_info("requested volume id out of range");
+ retcode = ERR_INVALID_REQUEST;
+ goto out;
+ }
-static unsigned short *
-tl_add_int(unsigned short *tl, enum drbd_tags tag, const void *val)
-{
- put_unaligned(tag, tl++);
- switch(tag_type(tag)) {
- case TT_INTEGER:
- put_unaligned(sizeof(int), tl++);
- put_unaligned(*(int *)val, (int *)tl);
- tl = (unsigned short*)((char*)tl+sizeof(int));
- break;
- case TT_INT64:
- put_unaligned(sizeof(u64), tl++);
- put_unaligned(*(u64 *)val, (u64 *)tl);
- tl = (unsigned short*)((char*)tl+sizeof(u64));
- break;
- default:
- /* someone did something stupid. */
- ;
+ /* drbd_adm_prepare made sure already
+ * that mdev->tconn and mdev->vnr match the request. */
+ if (adm_ctx.mdev) {
+ if (info->nlhdr->nlmsg_flags & NLM_F_EXCL)
+ retcode = ERR_MINOR_EXISTS;
+ /* else: still NO_ERROR */
+ goto out;
}
- return tl;
+
+ down_write(&drbd_cfg_rwsem);
+ retcode = conn_new_minor(adm_ctx.tconn, dh->minor, adm_ctx.volume);
+ up_write(&drbd_cfg_rwsem);
+out:
+ drbd_adm_finish(info, retcode);
+ return 0;
}
-void drbd_bcast_state(struct drbd_conf *mdev, union drbd_state state)
+static enum drbd_ret_code adm_delete_minor(struct drbd_conf *mdev)
{
- char buffer[sizeof(struct cn_msg)+
- sizeof(struct drbd_nl_cfg_reply)+
- sizeof(struct get_state_tag_len_struct)+
- sizeof(short int)];
- struct cn_msg *cn_reply = (struct cn_msg *) buffer;
- struct drbd_nl_cfg_reply *reply =
- (struct drbd_nl_cfg_reply *)cn_reply->data;
- unsigned short *tl = reply->tag_list;
-
- /* dev_warn(DEV, "drbd_bcast_state() got called\n"); */
-
- tl = get_state_to_tags(mdev, (struct get_state *)&state, tl);
-
- put_unaligned(TT_END, tl++); /* Close the tag list */
-
- cn_reply->id.idx = CN_IDX_DRBD;
- cn_reply->id.val = CN_VAL_DRBD;
-
- cn_reply->seq = atomic_add_return(1, &drbd_nl_seq);
- cn_reply->ack = 0; /* not used here. */
- cn_reply->len = sizeof(struct drbd_nl_cfg_reply) +
- (int)((char *)tl - (char *)reply->tag_list);
- cn_reply->flags = 0;
-
- reply->packet_type = P_get_state;
- reply->minor = mdev_to_minor(mdev);
- reply->ret_code = NO_ERROR;
-
- cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO);
+ if (mdev->state.disk == D_DISKLESS &&
+ /* no need to be mdev->state.conn == C_STANDALONE &&
+ * we may want to delete a minor from a live replication group.
+ */
+ mdev->state.role == R_SECONDARY) {
+ drbd_delete_device(mdev);
+ return NO_ERROR;
+ } else
+ return ERR_MINOR_CONFIGURED;
}
-void drbd_bcast_ev_helper(struct drbd_conf *mdev, char *helper_name)
+int drbd_adm_delete_minor(struct sk_buff *skb, struct genl_info *info)
{
- char buffer[sizeof(struct cn_msg)+
- sizeof(struct drbd_nl_cfg_reply)+
- sizeof(struct call_helper_tag_len_struct)+
- sizeof(short int)];
- struct cn_msg *cn_reply = (struct cn_msg *) buffer;
- struct drbd_nl_cfg_reply *reply =
- (struct drbd_nl_cfg_reply *)cn_reply->data;
- unsigned short *tl = reply->tag_list;
-
- /* dev_warn(DEV, "drbd_bcast_state() got called\n"); */
-
- tl = tl_add_str(tl, T_helper, helper_name);
- put_unaligned(TT_END, tl++); /* Close the tag list */
-
- cn_reply->id.idx = CN_IDX_DRBD;
- cn_reply->id.val = CN_VAL_DRBD;
-
- cn_reply->seq = atomic_add_return(1, &drbd_nl_seq);
- cn_reply->ack = 0; /* not used here. */
- cn_reply->len = sizeof(struct drbd_nl_cfg_reply) +
- (int)((char *)tl - (char *)reply->tag_list);
- cn_reply->flags = 0;
+ enum drbd_ret_code retcode;
- reply->packet_type = P_call_helper;
- reply->minor = mdev_to_minor(mdev);
- reply->ret_code = NO_ERROR;
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_MINOR);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
- cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO);
+ down_write(&drbd_cfg_rwsem);
+ retcode = adm_delete_minor(adm_ctx.mdev);
+ up_write(&drbd_cfg_rwsem);
+out:
+ drbd_adm_finish(info, retcode);
+ return 0;
}
-void drbd_bcast_ee(struct drbd_conf *mdev, const char *reason, const int dgs,
- const char *seen_hash, const char *calc_hash,
- const struct drbd_peer_request *peer_req)
+int drbd_adm_down(struct sk_buff *skb, struct genl_info *info)
{
- struct cn_msg *cn_reply;
- struct drbd_nl_cfg_reply *reply;
- unsigned short *tl;
- struct page *page;
- unsigned len;
+ enum drbd_ret_code retcode;
+ enum drbd_state_rv rv;
+ struct drbd_conf *mdev;
+ unsigned i;
- if (!peer_req)
- return;
- if (!reason || !reason[0])
- return;
+ retcode = drbd_adm_prepare(skb, info, 0);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
- /* apparently we have to memcpy twice, first to prepare the data for the
- * struct cn_msg, then within cn_netlink_send from the cn_msg to the
- * netlink skb. */
- /* receiver thread context, which is not in the writeout path (of this node),
- * but may be in the writeout path of the _other_ node.
- * GFP_NOIO to avoid potential "distributed deadlock". */
- cn_reply = kzalloc(
- sizeof(struct cn_msg)+
- sizeof(struct drbd_nl_cfg_reply)+
- sizeof(struct dump_ee_tag_len_struct)+
- sizeof(short int),
- GFP_NOIO);
-
- if (!cn_reply) {
- dev_err(DEV, "could not kmalloc buffer for drbd_bcast_ee, "
- "sector %llu, size %u\n",
- (unsigned long long)peer_req->i.sector,
- peer_req->i.size);
- return;
+ if (!adm_ctx.tconn) {
+ retcode = ERR_CONN_NOT_KNOWN;
+ goto out;
}
- reply = (struct drbd_nl_cfg_reply*)cn_reply->data;
- tl = reply->tag_list;
-
- tl = tl_add_str(tl, T_dump_ee_reason, reason);
- tl = tl_add_blob(tl, T_seen_digest, seen_hash, dgs);
- tl = tl_add_blob(tl, T_calc_digest, calc_hash, dgs);
- tl = tl_add_int(tl, T_ee_sector, &peer_req->i.sector);
- tl = tl_add_int(tl, T_ee_block_id, &peer_req->block_id);
-
- /* dump the first 32k */
- len = min_t(unsigned, peer_req->i.size, 32 << 10);
- put_unaligned(T_ee_data, tl++);
- put_unaligned(len, tl++);
-
- page = peer_req->pages;
- page_chain_for_each(page) {
- void *d = kmap_atomic(page, KM_USER0);
- unsigned l = min_t(unsigned, len, PAGE_SIZE);
- memcpy(tl, d, l);
- kunmap_atomic(d, KM_USER0);
- tl = (unsigned short*)((char*)tl + l);
- len -= l;
- if (len == 0)
- break;
+ down_read(&drbd_cfg_rwsem);
+ /* demote */
+ idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) {
+ retcode = drbd_set_role(mdev, R_SECONDARY, 0);
+ if (retcode < SS_SUCCESS) {
+ drbd_msg_put_info("failed to demote");
+ goto out_unlock;
+ }
}
- put_unaligned(TT_END, tl++); /* Close the tag list */
-
- cn_reply->id.idx = CN_IDX_DRBD;
- cn_reply->id.val = CN_VAL_DRBD;
-
- cn_reply->seq = atomic_add_return(1,&drbd_nl_seq);
- cn_reply->ack = 0; // not used here.
- cn_reply->len = sizeof(struct drbd_nl_cfg_reply) +
- (int)((char*)tl - (char*)reply->tag_list);
- cn_reply->flags = 0;
-
- reply->packet_type = P_dump_ee;
- reply->minor = mdev_to_minor(mdev);
- reply->ret_code = NO_ERROR;
-
- cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO);
- kfree(cn_reply);
-}
-void drbd_bcast_sync_progress(struct drbd_conf *mdev)
-{
- char buffer[sizeof(struct cn_msg)+
- sizeof(struct drbd_nl_cfg_reply)+
- sizeof(struct sync_progress_tag_len_struct)+
- sizeof(short int)];
- struct cn_msg *cn_reply = (struct cn_msg *) buffer;
- struct drbd_nl_cfg_reply *reply =
- (struct drbd_nl_cfg_reply *)cn_reply->data;
- unsigned short *tl = reply->tag_list;
- unsigned long rs_left;
- unsigned int res;
+ /* disconnect */
+ rv = conn_try_disconnect(adm_ctx.tconn, 0);
+ if (rv < SS_SUCCESS) {
+ retcode = rv; /* enum type mismatch! */
+ drbd_msg_put_info("failed to disconnect");
+ goto out_unlock;
+ }
- /* no local ref, no bitmap, no syncer progress, no broadcast. */
- if (!get_ldev(mdev))
- return;
- drbd_get_syncer_progress(mdev, &rs_left, &res);
- put_ldev(mdev);
+ /* detach */
+ idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) {
+ rv = adm_detach(mdev);
+ if (rv < SS_SUCCESS) {
+ retcode = rv; /* enum type mismatch! */
+ drbd_msg_put_info("failed to detach");
+ goto out_unlock;
+ }
+ }
+ up_read(&drbd_cfg_rwsem);
- tl = tl_add_int(tl, T_sync_progress, &res);
- put_unaligned(TT_END, tl++); /* Close the tag list */
+ /* delete volumes */
+ down_write(&drbd_cfg_rwsem);
+ idr_for_each_entry(&adm_ctx.tconn->volumes, mdev, i) {
+ retcode = adm_delete_minor(mdev);
+ if (retcode != NO_ERROR) {
+ /* "can not happen" */
+ drbd_msg_put_info("failed to delete volume");
+ up_write(&drbd_cfg_rwsem);
+ goto out;
+ }
+ }
- cn_reply->id.idx = CN_IDX_DRBD;
- cn_reply->id.val = CN_VAL_DRBD;
+ /* stop all threads */
+ conn_reconfig_done(adm_ctx.tconn);
- cn_reply->seq = atomic_add_return(1, &drbd_nl_seq);
- cn_reply->ack = 0; /* not used here. */
- cn_reply->len = sizeof(struct drbd_nl_cfg_reply) +
- (int)((char *)tl - (char *)reply->tag_list);
- cn_reply->flags = 0;
+ /* delete connection */
+ if (conn_lowest_minor(adm_ctx.tconn) < 0) {
+ list_del(&adm_ctx.tconn->all_tconn);
+ kref_put(&adm_ctx.tconn->kref, &conn_destroy);
- reply->packet_type = P_sync_progress;
- reply->minor = mdev_to_minor(mdev);
- reply->ret_code = NO_ERROR;
+ retcode = NO_ERROR;
+ } else {
+ /* "can not happen" */
+ retcode = ERR_CONN_IN_USE;
+ drbd_msg_put_info("failed to delete connection");
+ }
- cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO);
+ up_write(&drbd_cfg_rwsem);
+ goto out;
+out_unlock:
+ up_read(&drbd_cfg_rwsem);
+out:
+ drbd_adm_finish(info, retcode);
+ return 0;
}
-int __init drbd_nl_init(void)
+int drbd_adm_delete_connection(struct sk_buff *skb, struct genl_info *info)
{
- static struct cb_id cn_id_drbd;
- int err, try=10;
+ enum drbd_ret_code retcode;
- cn_id_drbd.val = CN_VAL_DRBD;
- do {
- cn_id_drbd.idx = cn_idx;
- err = cn_add_callback(&cn_id_drbd, "cn_drbd", &drbd_connector_callback);
- if (!err)
- break;
- cn_idx = (cn_idx + CN_IDX_STEP);
- } while (try--);
+ retcode = drbd_adm_prepare(skb, info, DRBD_ADM_NEED_CONN);
+ if (!adm_ctx.reply_skb)
+ return retcode;
+ if (retcode != NO_ERROR)
+ goto out;
- if (err) {
- printk(KERN_ERR "drbd: cn_drbd failed to register\n");
- return err;
+ down_write(&drbd_cfg_rwsem);
+ if (conn_lowest_minor(adm_ctx.tconn) < 0) {
+ list_del(&adm_ctx.tconn->all_tconn);
+ kref_put(&adm_ctx.tconn->kref, &conn_destroy);
+
+ retcode = NO_ERROR;
+ } else {
+ retcode = ERR_CONN_IN_USE;
}
+ up_write(&drbd_cfg_rwsem);
+out:
+ drbd_adm_finish(info, retcode);
return 0;
}
-void drbd_nl_cleanup(void)
-{
- static struct cb_id cn_id_drbd;
-
- cn_id_drbd.idx = cn_idx;
- cn_id_drbd.val = CN_VAL_DRBD;
-
- cn_del_callback(&cn_id_drbd);
-}
-
-void drbd_nl_send_reply(struct cn_msg *req, int ret_code)
+void drbd_bcast_event(struct drbd_conf *mdev, const struct sib_info *sib)
{
- char buffer[sizeof(struct cn_msg)+sizeof(struct drbd_nl_cfg_reply)];
- struct cn_msg *cn_reply = (struct cn_msg *) buffer;
- struct drbd_nl_cfg_reply *reply =
- (struct drbd_nl_cfg_reply *)cn_reply->data;
- int rr;
+ static atomic_t drbd_genl_seq = ATOMIC_INIT(2); /* two. */
+ struct sk_buff *msg;
+ struct drbd_genlmsghdr *d_out;
+ unsigned seq;
+ int err = -ENOMEM;
+
+ seq = atomic_inc_return(&drbd_genl_seq);
+ msg = genlmsg_new(NLMSG_GOODSIZE, GFP_NOIO);
+ if (!msg)
+ goto failed;
+
+ err = -EMSGSIZE;
+ d_out = genlmsg_put(msg, 0, seq, &drbd_genl_family, 0, DRBD_EVENT);
+ if (!d_out) /* cannot happen, but anyways. */
+ goto nla_put_failure;
+ d_out->minor = mdev_to_minor(mdev);
+ d_out->ret_code = 0;
+
+ if (nla_put_status_info(msg, mdev, sib))
+ goto nla_put_failure;
+ genlmsg_end(msg, d_out);
+ err = drbd_genl_multicast_events(msg, 0);
+ /* msg has been consumed or freed in netlink_broadcast() */
+ if (err && err != -ESRCH)
+ goto failed;
- memset(buffer, 0, sizeof(buffer));
- cn_reply->id = req->id;
-
- cn_reply->seq = req->seq;
- cn_reply->ack = req->ack + 1;
- cn_reply->len = sizeof(struct drbd_nl_cfg_reply);
- cn_reply->flags = 0;
-
- reply->packet_type = P_return_code_only;
- reply->minor = ((struct drbd_nl_cfg_req *)req->data)->drbd_minor;
- reply->ret_code = ret_code;
+ return;
- rr = cn_netlink_send(cn_reply, CN_IDX_DRBD, GFP_NOIO);
- if (rr && rr != -ESRCH)
- printk(KERN_INFO "drbd: cn_netlink_send()=%d\n", rr);
+nla_put_failure:
+ nlmsg_free(msg);
+failed:
+ dev_err(DEV, "Error %d while broadcasting event. "
+ "Event seq:%u sib_reason:%u\n",
+ err, seq, sib->sib_reason);
}
-