Whamcloud - gitweb
LU-13004 ptlrpc: Allow BULK_BUF_KIOV to accept a kvec
[fs/lustre-release.git] / lustre / ptlrpc / client.c
index e43cd01..85bb30d 100644 (file)
@@ -37,6 +37,7 @@
 #include <linux/delay.h>
 #include <linux/random.h>
 
+#include <lnet/lib-lnet.h>
 #include <obd_support.h>
 #include <obd_class.h>
 #include <lustre_lib.h>
@@ -68,6 +69,28 @@ static void ptlrpc_release_bulk_page_pin(struct ptlrpc_bulk_desc *desc)
                put_page(BD_GET_KIOV(desc, i).kiov_page);
 }
 
+static int ptlrpc_prep_bulk_frag_pages(struct ptlrpc_bulk_desc *desc,
+                                      void *frag, int len)
+{
+       unsigned int offset = (uintptr_t)frag & ~PAGE_MASK;
+
+       ENTRY;
+       while (len > 0) {
+               int page_len = min_t(unsigned int, PAGE_SIZE - offset,
+                                    len);
+               uintptr_t vaddr = (uintptr_t) frag;
+
+               ptlrpc_prep_bulk_page_nopin(desc,
+                                           lnet_kvaddr_to_page(vaddr),
+                                           offset, page_len);
+               offset = 0;
+               len -= page_len;
+               frag += page_len;
+       }
+
+       RETURN(desc->bd_nob);
+}
+
 const struct ptlrpc_bulk_frag_ops ptlrpc_bulk_kiov_pin_ops = {
        .add_kiov_frag  = ptlrpc_prep_bulk_page_pin,
        .release_frags  = ptlrpc_release_bulk_page_pin,
@@ -77,6 +100,7 @@ EXPORT_SYMBOL(ptlrpc_bulk_kiov_pin_ops);
 const struct ptlrpc_bulk_frag_ops ptlrpc_bulk_kiov_nopin_ops = {
        .add_kiov_frag  = ptlrpc_prep_bulk_page_nopin,
        .release_frags  = ptlrpc_release_bulk_noop,
+       .add_iov_frag   = ptlrpc_prep_bulk_frag_pages,
 };
 EXPORT_SYMBOL(ptlrpc_bulk_kiov_nopin_ops);
 
@@ -217,7 +241,6 @@ struct ptlrpc_bulk_desc *ptlrpc_prep_bulk_imp(struct ptlrpc_request *req,
        if (!desc)
                RETURN(NULL);
 
-       desc->bd_import_generation = req->rq_import_generation;
        desc->bd_import = class_import_get(imp);
        desc->bd_req = req;
 
@@ -437,14 +460,16 @@ static int unpack_reply(struct ptlrpc_request *req)
        if (SPTLRPC_FLVR_POLICY(req->rq_flvr.sf_rpc) != SPTLRPC_POLICY_NULL) {
                rc = ptlrpc_unpack_rep_msg(req, req->rq_replen);
                if (rc) {
-                       DEBUG_REQ(D_ERROR, req, "unpack_rep failed: %d", rc);
+                       DEBUG_REQ(D_ERROR, req, "unpack_rep failed: rc = %d",
+                                 rc);
                        return -EPROTO;
                }
        }
 
        rc = lustre_unpack_rep_ptlrpc_body(req, MSG_PTLRPC_BODY_OFF);
        if (rc) {
-               DEBUG_REQ(D_ERROR, req, "unpack ptlrpc body failed: %d", rc);
+               DEBUG_REQ(D_ERROR, req, "unpack ptlrpc body failed: rc = %d",
+                         rc);
                return -EPROTO;
        }
        return 0;
@@ -505,6 +530,8 @@ __must_hold(&req->rq_lock)
        req->rq_deadline = req->rq_sent + req->rq_timeout +
                           ptlrpc_at_get_net_latency(req);
 
+       /* The below message is checked in replay-single.sh test_65{a,b} */
+       /* The below message is checked in sanity-{gss,krb5} test_8 */
        DEBUG_REQ(D_ADAPTTO, req,
                  "Early reply #%d, new deadline in %llds (%llds)",
                  req->rq_early_count,
@@ -621,7 +648,7 @@ ptlrpc_init_rq_pool(int num_rq, int msgsize,
 {
        struct ptlrpc_request_pool *pool;
 
-       OBD_ALLOC(pool, sizeof(struct ptlrpc_request_pool));
+       OBD_ALLOC_PTR(pool);
        if (!pool)
                return NULL;
 
@@ -734,6 +761,41 @@ static inline void ptlrpc_assign_next_xid(struct ptlrpc_request *req)
 
 static atomic64_t ptlrpc_last_xid;
 
+static void ptlrpc_reassign_next_xid(struct ptlrpc_request *req)
+{
+       spin_lock(&req->rq_import->imp_lock);
+       list_del_init(&req->rq_unreplied_list);
+       ptlrpc_assign_next_xid_nolock(req);
+       spin_unlock(&req->rq_import->imp_lock);
+       DEBUG_REQ(D_RPCTRACE, req, "reassign xid");
+}
+
+void ptlrpc_get_mod_rpc_slot(struct ptlrpc_request *req)
+{
+       struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
+       __u32 opc;
+       __u16 tag;
+
+       opc = lustre_msg_get_opc(req->rq_reqmsg);
+       tag = obd_get_mod_rpc_slot(cli, opc);
+       lustre_msg_set_tag(req->rq_reqmsg, tag);
+       ptlrpc_reassign_next_xid(req);
+}
+EXPORT_SYMBOL(ptlrpc_get_mod_rpc_slot);
+
+void ptlrpc_put_mod_rpc_slot(struct ptlrpc_request *req)
+{
+       __u16 tag = lustre_msg_get_tag(req->rq_reqmsg);
+
+       if (tag != 0) {
+               struct client_obd *cli = &req->rq_import->imp_obd->u.cli;
+               __u32 opc = lustre_msg_get_opc(req->rq_reqmsg);
+
+               obd_put_mod_rpc_slot(cli, opc, tag);
+       }
+}
+EXPORT_SYMBOL(ptlrpc_put_mod_rpc_slot);
+
 int ptlrpc_request_bufs_pack(struct ptlrpc_request *request,
                             __u32 version, int opcode, char **bufs,
                             struct ptlrpc_cli_ctx *ctx)
@@ -841,32 +903,7 @@ EXPORT_SYMBOL(ptlrpc_request_bufs_pack);
 int ptlrpc_request_pack(struct ptlrpc_request *request,
                        __u32 version, int opcode)
 {
-       int rc;
-
-       rc = ptlrpc_request_bufs_pack(request, version, opcode, NULL, NULL);
-       if (rc)
-               return rc;
-
-       /*
-        * For some old 1.8 clients (< 1.8.7), they will LASSERT the size of
-        * ptlrpc_body sent from server equal to local ptlrpc_body size, so we
-        * have to send old ptlrpc_body to keep interoprability with these
-        * clients.
-        *
-        * Only three kinds of server->client RPCs so far:
-        *  - LDLM_BL_CALLBACK
-        *  - LDLM_CP_CALLBACK
-        *  - LDLM_GL_CALLBACK
-        *
-        * XXX This should be removed whenever we drop the interoprability with
-        *     the these old clients.
-        */
-       if (opcode == LDLM_BL_CALLBACK || opcode == LDLM_CP_CALLBACK ||
-           opcode == LDLM_GL_CALLBACK)
-               req_capsule_shrink(&request->rq_pill, &RMF_PTLRPC_BODY,
-                                  sizeof(struct ptlrpc_body_v2), RCL_CLIENT);
-
-       return rc;
+       return ptlrpc_request_bufs_pack(request, version, opcode, NULL, NULL);
 }
 EXPORT_SYMBOL(ptlrpc_request_pack);
 
@@ -916,7 +953,6 @@ ptlrpc_request_alloc_internal(struct obd_import *imp,
                              const struct req_format *format)
 {
        struct ptlrpc_request *request;
-       int connect = 0;
 
        request = __ptlrpc_request_alloc(imp, pool);
        if (!request)
@@ -937,17 +973,17 @@ ptlrpc_request_alloc_internal(struct obd_import *imp,
                if (imp->imp_state == LUSTRE_IMP_IDLE) {
                        imp->imp_generation++;
                        imp->imp_initiated_at = imp->imp_generation;
-                       imp->imp_state =  LUSTRE_IMP_NEW;
-                       connect = 1;
-               }
-               spin_unlock(&imp->imp_lock);
-               if (connect) {
-                       rc = ptlrpc_connect_import(imp);
+                       imp->imp_state = LUSTRE_IMP_NEW;
+
+                       /* connect_import_locked releases imp_lock */
+                       rc = ptlrpc_connect_import_locked(imp);
                        if (rc < 0) {
                                ptlrpc_request_free(request);
                                return NULL;
                        }
                        ptlrpc_pinger_add_import(imp);
+               } else {
+                       spin_unlock(&imp->imp_lock);
                }
        }
 
@@ -1225,7 +1261,7 @@ static int ptlrpc_import_delay_req(struct obd_import *imp,
        if (req->rq_ctx_init || req->rq_ctx_fini) {
                /* always allow ctx init/fini rpc go through */
        } else if (imp->imp_state == LUSTRE_IMP_NEW) {
-               DEBUG_REQ(D_ERROR, req, "Uninitialized import.");
+               DEBUG_REQ(D_ERROR, req, "Uninitialized import");
                *status = -EIO;
        } else if (imp->imp_state == LUSTRE_IMP_CLOSED) {
                unsigned int opc = lustre_msg_get_opc(req->rq_reqmsg);
@@ -1235,11 +1271,11 @@ static int ptlrpc_import_delay_req(struct obd_import *imp,
                 * race with umount
                 */
                DEBUG_REQ((opc == OBD_PING || opc == OST_STATFS) ?
-                         D_HA : D_ERROR, req, "IMP_CLOSED ");
+                         D_HA : D_ERROR, req, "IMP_CLOSED");
                *status = -EIO;
        } else if (ptlrpc_send_limit_expired(req)) {
                /* probably doesn't need to be a D_ERROR afterinitial testing */
-               DEBUG_REQ(D_HA, req, "send limit expired ");
+               DEBUG_REQ(D_HA, req, "send limit expired");
                *status = -ETIMEDOUT;
        } else if (req->rq_send_state == LUSTRE_IMP_CONNECTING &&
                   imp->imp_state == LUSTRE_IMP_CONNECTING) {
@@ -1269,7 +1305,7 @@ static int ptlrpc_import_delay_req(struct obd_import *imp,
                            imp->imp_state == LUSTRE_IMP_REPLAY_LOCKS ||
                            imp->imp_state == LUSTRE_IMP_REPLAY_WAIT ||
                            imp->imp_state == LUSTRE_IMP_RECOVER)) {
-                       DEBUG_REQ(D_HA, req, "allow during recovery.\n");
+                       DEBUG_REQ(D_HA, req, "allow during recovery");
                } else {
                        delay = 1;
                }
@@ -1326,32 +1362,28 @@ static bool ptlrpc_console_allow(struct ptlrpc_request *req, __u32 opc, int err)
  */
 static int ptlrpc_check_status(struct ptlrpc_request *req)
 {
-       int err;
+       int rc;
 
        ENTRY;
-       err = lustre_msg_get_status(req->rq_repmsg);
+       rc = lustre_msg_get_status(req->rq_repmsg);
        if (lustre_msg_get_type(req->rq_repmsg) == PTL_RPC_MSG_ERR) {
                struct obd_import *imp = req->rq_import;
                lnet_nid_t nid = imp->imp_connection->c_peer.nid;
                __u32 opc = lustre_msg_get_opc(req->rq_reqmsg);
 
-               if (ptlrpc_console_allow(req, opc, err))
+               if (ptlrpc_console_allow(req, opc, rc))
                        LCONSOLE_ERROR_MSG(0x11,
                                           "%s: operation %s to node %s failed: rc = %d\n",
                                           imp->imp_obd->obd_name,
                                           ll_opcode2str(opc),
-                                          libcfs_nid2str(nid), err);
-               RETURN(err < 0 ? err : -EINVAL);
+                                          libcfs_nid2str(nid), rc);
+               RETURN(rc < 0 ? rc : -EINVAL);
        }
 
-       if (err < 0) {
-               DEBUG_REQ(D_INFO, req, "status is %d", err);
-       } else if (err > 0) {
-               /* XXX: translate this error from net to host */
-               DEBUG_REQ(D_INFO, req, "status is %d", err);
-       }
+       if (rc)
+               DEBUG_REQ(D_INFO, req, "check status: rc = %d", rc);
 
-       RETURN(err);
+       RETURN(rc);
 }
 
 /**
@@ -1419,7 +1451,7 @@ static int after_reply(struct ptlrpc_request *req)
        if (req->rq_reply_truncated) {
                if (ptlrpc_no_resend(req)) {
                        DEBUG_REQ(D_ERROR, req,
-                                 "reply buffer overflow, expected: %d, actual size: %d",
+                                 "reply buffer overflow, expected=%d, actual size=%d",
                                  req->rq_nob_received, req->rq_repbuf_len);
                        RETURN(-EOVERFLOW);
                }
@@ -1448,7 +1480,7 @@ static int after_reply(struct ptlrpc_request *req)
         */
        rc = sptlrpc_cli_unwrap_reply(req);
        if (rc) {
-               DEBUG_REQ(D_ERROR, req, "unwrap reply failed (%d):", rc);
+               DEBUG_REQ(D_ERROR, req, "unwrap reply failed: rc = %d", rc);
                RETURN(rc);
        }
 
@@ -1467,8 +1499,8 @@ static int after_reply(struct ptlrpc_request *req)
            ptlrpc_no_resend(req) == 0 && !req->rq_no_retry_einprogress) {
                time64_t now = ktime_get_real_seconds();
 
-               DEBUG_REQ(req->rq_nr_resend > 0 ? D_ERROR : D_RPCTRACE, req,
-                         "Resending request on EINPROGRESS");
+               DEBUG_REQ((req->rq_nr_resend % 8 == 1 ? D_WARNING : 0) |
+                         D_RPCTRACE, req, "resending request on EINPROGRESS");
                spin_lock(&req->rq_lock);
                req->rq_resend = 1;
                spin_unlock(&req->rq_lock);
@@ -1715,7 +1747,8 @@ static int ptlrpc_send_new_req(struct ptlrpc_request *req)
                RETURN(rc);
        }
        if (rc) {
-               DEBUG_REQ(D_HA, req, "send failed (%d); expect timeout", rc);
+               DEBUG_REQ(D_HA, req, "send failed, expect timeout: rc = %d",
+                         rc);
                spin_lock(&req->rq_lock);
                req->rq_net_err = 1;
                spin_unlock(&req->rq_lock);
@@ -3079,7 +3112,7 @@ static int ptlrpc_replay_interpret(const struct lu_env *env,
        if (!ptlrpc_client_replied(req) ||
            (req->rq_bulk &&
             lustre_msg_get_status(req->rq_repmsg) == -ETIMEDOUT)) {
-               DEBUG_REQ(D_ERROR, req, "request replay timed out.\n");
+               DEBUG_REQ(D_ERROR, req, "request replay timed out");
                GOTO(out, rc = -ETIMEDOUT);
        }
 
@@ -3091,7 +3124,7 @@ static int ptlrpc_replay_interpret(const struct lu_env *env,
        /** VBR: check version failure */
        if (lustre_msg_get_status(req->rq_repmsg) == -EOVERFLOW) {
                /** replay was failed due to version mismatch */
-               DEBUG_REQ(D_WARNING, req, "Version mismatch during replay\n");
+               DEBUG_REQ(D_WARNING, req, "Version mismatch during replay");
                spin_lock(&imp->imp_lock);
                imp->imp_vbr_failed = 1;
                spin_unlock(&imp->imp_lock);
@@ -3114,13 +3147,13 @@ static int ptlrpc_replay_interpret(const struct lu_env *env,
        /* transaction number shouldn't be bigger than the latest replayed */
        if (req->rq_transno > lustre_msg_get_transno(req->rq_reqmsg)) {
                DEBUG_REQ(D_ERROR, req,
-                         "Reported transno %llu is bigger than the replayed one: %llu",
+                         "Reported transno=%llu is bigger than replayed=%llu",
                          req->rq_transno,
                          lustre_msg_get_transno(req->rq_reqmsg));
                GOTO(out, rc = -EINVAL);
        }
 
-       DEBUG_REQ(D_HA, req, "got rep");
+       DEBUG_REQ(D_HA, req, "got reply");
 
        /* let the callback do fixups, possibly including in the request */
        if (req->rq_replay_cb)
@@ -3361,7 +3394,8 @@ void ptlrpc_init_xid(void)
        }
 
        /* Need to always be aligned to a power-of-two for mutli-bulk BRW */
-       CLASSERT((PTLRPC_BULK_OPS_COUNT & (PTLRPC_BULK_OPS_COUNT - 1)) == 0);
+       BUILD_BUG_ON((PTLRPC_BULK_OPS_COUNT & (PTLRPC_BULK_OPS_COUNT - 1)) !=
+                    0);
        xid &= PTLRPC_BULK_OPS_MASK;
        atomic64_set(&ptlrpc_last_xid, xid);
 }