drivers/infiniband/hw/ehca/ehca_irq.c

   1 /*
   2  *  IBM eServer eHCA Infiniband device driver for Linux on POWER
   3  *
   4  *  Functions for EQs, NEQs and interrupts
   5  *
   6  *  Authors: Heiko J Schick <schickhj@de.ibm.com>
   7  *           Khadija Souissi <souissi@de.ibm.com>
   8  *           Hoang-Nam Nguyen <hnguyen@de.ibm.com>
   9  *           Joachim Fenkes <fenkes@de.ibm.com>
  10  *
  11  *  Copyright (c) 2005 IBM Corporation
  12  *
  13  *  All rights reserved.
  14  *
  15  *  This source code is distributed under a dual license of GPL v2.0 and OpenIB
  16  *  BSD.
  17  *
  18  * OpenIB BSD License
  19  *
  20  * Redistribution and use in source and binary forms, with or without
  21  * modification, are permitted provided that the following conditions are met:
  22  *
  23  * Redistributions of source code must retain the above copyright notice, this
  24  * list of conditions and the following disclaimer.
  25  *
  26  * Redistributions in binary form must reproduce the above copyright notice,
  27  * this list of conditions and the following disclaimer in the documentation
  28  * and/or other materials
  29  * provided with the distribution.
  30  *
  31  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
  32  * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  33  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  34  * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
  35  * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
  36  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
  37  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
  38  * BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
  39  * IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
  40  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
  41  * POSSIBILITY OF SUCH DAMAGE.
  42  */
  43
  44 #include "ehca_classes.h"
  45 #include "ehca_irq.h"
  46 #include "ehca_iverbs.h"
  47 #include "ehca_tools.h"
  48 #include "hcp_if.h"
  49 #include "hipz_fns.h"
  50 #include "ipz_pt_fn.h"
  51
  52 #define EQE_COMPLETION_EVENT   EHCA_BMASK_IBM( 1,  1)
  53 #define EQE_CQ_QP_NUMBER       EHCA_BMASK_IBM( 8, 31)
  54 #define EQE_EE_IDENTIFIER      EHCA_BMASK_IBM( 2,  7)
  55 #define EQE_CQ_NUMBER          EHCA_BMASK_IBM( 8, 31)
  56 #define EQE_QP_NUMBER          EHCA_BMASK_IBM( 8, 31)
  57 #define EQE_QP_TOKEN           EHCA_BMASK_IBM(32, 63)
  58 #define EQE_CQ_TOKEN           EHCA_BMASK_IBM(32, 63)
  59
  60 #define NEQE_COMPLETION_EVENT  EHCA_BMASK_IBM( 1,  1)
  61 #define NEQE_EVENT_CODE        EHCA_BMASK_IBM( 2,  7)
  62 #define NEQE_PORT_NUMBER       EHCA_BMASK_IBM( 8, 15)
  63 #define NEQE_PORT_AVAILABILITY EHCA_BMASK_IBM(16, 16)
  64 #define NEQE_DISRUPTIVE        EHCA_BMASK_IBM(16, 16)
  65 #define NEQE_SPECIFIC_EVENT    EHCA_BMASK_IBM(16, 23)
  66
  67 #define ERROR_DATA_LENGTH      EHCA_BMASK_IBM(52, 63)
  68 #define ERROR_DATA_TYPE        EHCA_BMASK_IBM( 0,  7)
  69
  70 static void queue_comp_task(struct ehca_cq *__cq);
  71
  72 static struct ehca_comp_pool *pool;
  73
  74 static inline void comp_event_callback(struct ehca_cq *cq)
  75 {
  76         if (!cq->ib_cq.comp_handler)
  77                 return;
  78
  79         spin_lock(&cq->cb_lock);
  80         cq->ib_cq.comp_handler(&cq->ib_cq, cq->ib_cq.cq_context);
  81         spin_unlock(&cq->cb_lock);
  82
  83         return;
  84 }
  85
  86 static void print_error_data(struct ehca_shca *shca, void *data,
  87                              u64 *rblock, int length)
  88 {
  89         u64 type = EHCA_BMASK_GET(ERROR_DATA_TYPE, rblock[2]);
  90         u64 resource = rblock[1];
  91
  92         switch (type) {
  93         case 0x1: /* Queue Pair */
  94         {
  95                 struct ehca_qp *qp = (struct ehca_qp *)data;
  96
  97                 /* only print error data if AER is set */
  98                 if (rblock[6] == 0)
  99                         return;
 100
 101                 ehca_err(&shca->ib_device,
 102                          "QP 0x%x (resource=%lx) has errors.",
 103                          qp->ib_qp.qp_num, resource);
 104                 break;
 105         }
 106         case 0x4: /* Completion Queue */
 107         {
 108                 struct ehca_cq *cq = (struct ehca_cq *)data;
 109
 110                 ehca_err(&shca->ib_device,
 111                          "CQ 0x%x (resource=%lx) has errors.",
 112                          cq->cq_number, resource);
 113                 break;
 114         }
 115         default:
 116                 ehca_err(&shca->ib_device,
 117                          "Unknown error type: %lx on %s.",
 118                          type, shca->ib_device.name);
 119                 break;
 120         }
 121
 122         ehca_err(&shca->ib_device, "Error data is available: %lx.", resource);
 123         ehca_err(&shca->ib_device, "EHCA ----- error data begin "
 124                  "---------------------------------------------------");
 125         ehca_dmp(rblock, length, "resource=%lx", resource);
 126         ehca_err(&shca->ib_device, "EHCA ----- error data end "
 127                  "----------------------------------------------------");
 128
 129         return;
 130 }
 131
 132 int ehca_error_data(struct ehca_shca *shca, void *data,
 133                     u64 resource)
 134 {
 135
 136         unsigned long ret;
 137         u64 *rblock;
 138         unsigned long block_count;
 139
 140         rblock = ehca_alloc_fw_ctrlblock(GFP_ATOMIC);
 141         if (!rblock) {
 142                 ehca_err(&shca->ib_device, "Cannot allocate rblock memory.");
 143                 ret = -ENOMEM;
 144                 goto error_data1;
 145         }
 146
 147         /* rblock must be 4K aligned and should be 4K large */
 148         ret = hipz_h_error_data(shca->ipz_hca_handle,
 149                                 resource,
 150                                 rblock,
 151                                 &block_count);
 152
 153         if (ret == H_R_STATE)
 154                 ehca_err(&shca->ib_device,
 155                          "No error data is available: %lx.", resource);
 156         else if (ret == H_SUCCESS) {
 157                 int length;
 158
 159                 length = EHCA_BMASK_GET(ERROR_DATA_LENGTH, rblock[0]);
 160
 161                 if (length > EHCA_PAGESIZE)
 162                         length = EHCA_PAGESIZE;
 163
 164                 print_error_data(shca, data, rblock, length);
 165         } else
 166                 ehca_err(&shca->ib_device,
 167                          "Error data could not be fetched: %lx", resource);
 168
 169         ehca_free_fw_ctrlblock(rblock);
 170
 171 error_data1:
 172         return ret;
 173
 174 }
 175
 176 static void dispatch_qp_event(struct ehca_shca *shca, struct ehca_qp *qp,
 177                               enum ib_event_type event_type)
 178 {
 179         struct ib_event event;
 180
 181         /* PATH_MIG without the QP ever having been armed is false alarm */
 182         if (event_type == IB_EVENT_PATH_MIG && !qp->mig_armed)
 183                 return;
 184
 185         event.device = &shca->ib_device;
 186         event.event = event_type;
 187
 188         if (qp->ext_type == EQPT_SRQ) {
 189                 if (!qp->ib_srq.event_handler)
 190                         return;
 191
 192                 event.element.srq = &qp->ib_srq;
 193                 qp->ib_srq.event_handler(&event, qp->ib_srq.srq_context);
 194         } else {
 195                 if (!qp->ib_qp.event_handler)
 196                         return;
 197
 198                 event.element.qp = &qp->ib_qp;
 199                 qp->ib_qp.event_handler(&event, qp->ib_qp.qp_context);
 200         }
 201 }
 202
 203 static void qp_event_callback(struct ehca_shca *shca, u64 eqe,
 204                               enum ib_event_type event_type, int fatal)
 205 {
 206         struct ehca_qp *qp;
 207         u32 token = EHCA_BMASK_GET(EQE_QP_TOKEN, eqe);
 208
 209         read_lock(&ehca_qp_idr_lock);
 210         qp = idr_find(&ehca_qp_idr, token);
 211         if (qp)
 212                 atomic_inc(&qp->nr_events);
 213         read_unlock(&ehca_qp_idr_lock);
 214
 215         if (!qp)
 216                 return;
 217
 218         if (fatal)
 219                 ehca_error_data(shca, qp, qp->ipz_qp_handle.handle);
 220
 221         dispatch_qp_event(shca, qp, fatal && qp->ext_type == EQPT_SRQ ?
 222                           IB_EVENT_SRQ_ERR : event_type);
 223
 224         /*
 225          * eHCA only processes one WQE at a time for SRQ base QPs,
 226          * so the last WQE has been processed as soon as the QP enters
 227          * error state.
 228          */
 229         if (fatal && qp->ext_type == EQPT_SRQBASE)
 230                 dispatch_qp_event(shca, qp, IB_EVENT_QP_LAST_WQE_REACHED);
 231
 232         if (atomic_dec_and_test(&qp->nr_events))
 233                 wake_up(&qp->wait_completion);
 234         return;
 235 }
 236
 237 static void cq_event_callback(struct ehca_shca *shca,
 238                               u64 eqe)
 239 {
 240         struct ehca_cq *cq;
 241         u32 token = EHCA_BMASK_GET(EQE_CQ_TOKEN, eqe);
 242
 243         read_lock(&ehca_cq_idr_lock);
 244         cq = idr_find(&ehca_cq_idr, token);
 245         if (cq)
 246                 atomic_inc(&cq->nr_events);
 247         read_unlock(&ehca_cq_idr_lock);
 248
 249         if (!cq)
 250                 return;
 251
 252         ehca_error_data(shca, cq, cq->ipz_cq_handle.handle);
 253
 254         if (atomic_dec_and_test(&cq->nr_events))
 255                 wake_up(&cq->wait_completion);
 256
 257         return;
 258 }
 259
 260 static void parse_identifier(struct ehca_shca *shca, u64 eqe)
 261 {
 262         u8 identifier = EHCA_BMASK_GET(EQE_EE_IDENTIFIER, eqe);
 263
 264         switch (identifier) {
 265         case 0x02: /* path migrated */
 266                 qp_event_callback(shca, eqe, IB_EVENT_PATH_MIG, 0);
 267                 break;
 268         case 0x03: /* communication established */
 269                 qp_event_callback(shca, eqe, IB_EVENT_COMM_EST, 0);
 270                 break;
 271         case 0x04: /* send queue drained */
 272                 qp_event_callback(shca, eqe, IB_EVENT_SQ_DRAINED, 0);
 273                 break;
 274         case 0x05: /* QP error */
 275         case 0x06: /* QP error */
 276                 qp_event_callback(shca, eqe, IB_EVENT_QP_FATAL, 1);
 277                 break;
 278         case 0x07: /* CQ error */
 279         case 0x08: /* CQ error */
 280                 cq_event_callback(shca, eqe);
 281                 break;
 282         case 0x09: /* MRMWPTE error */
 283                 ehca_err(&shca->ib_device, "MRMWPTE error.");
 284                 break;
 285         case 0x0A: /* port event */
 286                 ehca_err(&shca->ib_device, "Port event.");
 287                 break;
 288         case 0x0B: /* MR access error */
 289                 ehca_err(&shca->ib_device, "MR access error.");
 290                 break;
 291         case 0x0C: /* EQ error */
 292                 ehca_err(&shca->ib_device, "EQ error.");
 293                 break;
 294         case 0x0D: /* P/Q_Key mismatch */
 295                 ehca_err(&shca->ib_device, "P/Q_Key mismatch.");
 296                 break;
 297         case 0x10: /* sampling complete */
 298                 ehca_err(&shca->ib_device, "Sampling complete.");
 299                 break;
 300         case 0x11: /* unaffiliated access error */
 301                 ehca_err(&shca->ib_device, "Unaffiliated access error.");
 302                 break;
 303         case 0x12: /* path migrating */
 304                 ehca_err(&shca->ib_device, "Path migrating.");
 305                 break;
 306         case 0x13: /* interface trace stopped */
 307                 ehca_err(&shca->ib_device, "Interface trace stopped.");
 308                 break;
 309         case 0x14: /* first error capture info available */
 310                 ehca_info(&shca->ib_device, "First error capture available");
 311                 break;
 312         case 0x15: /* SRQ limit reached */
 313                 qp_event_callback(shca, eqe, IB_EVENT_SRQ_LIMIT_REACHED, 0);
 314                 break;
 315         default:
 316                 ehca_err(&shca->ib_device, "Unknown identifier: %x on %s.",
 317                          identifier, shca->ib_device.name);
 318                 break;
 319         }
 320
 321         return;
 322 }
 323
 324 static void dispatch_port_event(struct ehca_shca *shca, int port_num,
 325                                 enum ib_event_type type, const char *msg)
 326 {
 327         struct ib_event event;
 328
 329         ehca_info(&shca->ib_device, "port %d %s.", port_num, msg);
 330         event.device = &shca->ib_device;
 331         event.event = type;
 332         event.element.port_num = port_num;
 333         ib_dispatch_event(&event);
 334 }
 335
 336 static void notify_port_conf_change(struct ehca_shca *shca, int port_num)
 337 {
 338         struct ehca_sma_attr  new_attr;
 339         struct ehca_sma_attr *old_attr = &shca->sport[port_num - 1].saved_attr;
 340
 341         ehca_query_sma_attr(shca, port_num, &new_attr);
 342
 343         if (new_attr.sm_sl  != old_attr->sm_sl ||
 344             new_attr.sm_lid != old_attr->sm_lid)
 345                 dispatch_port_event(shca, port_num, IB_EVENT_SM_CHANGE,
 346                                     "SM changed");
 347
 348         if (new_attr.lid != old_attr->lid ||
 349             new_attr.lmc != old_attr->lmc)
 350                 dispatch_port_event(shca, port_num, IB_EVENT_LID_CHANGE,
 351                                     "LID changed");
 352
 353         if (new_attr.pkey_tbl_len != old_attr->pkey_tbl_len ||
 354             memcmp(new_attr.pkeys, old_attr->pkeys,
 355                    sizeof(u16) * new_attr.pkey_tbl_len))
 356                 dispatch_port_event(shca, port_num, IB_EVENT_PKEY_CHANGE,
 357                                     "P_Key changed");
 358
 359         *old_attr = new_attr;
 360 }
 361
 362 static void parse_ec(struct ehca_shca *shca, u64 eqe)
 363 {
 364         u8 ec   = EHCA_BMASK_GET(NEQE_EVENT_CODE, eqe);
 365         u8 port = EHCA_BMASK_GET(NEQE_PORT_NUMBER, eqe);
 366         u8 spec_event;
 367         struct ehca_sport *sport = &shca->sport[port - 1];
 368         unsigned long flags;
 369
 370         switch (ec) {
 371         case 0x30: /* port availability change */
 372                 if (EHCA_BMASK_GET(NEQE_PORT_AVAILABILITY, eqe)) {
 373                         int suppress_event;
 374                         /* replay modify_qp for sqps */
 375                         spin_lock_irqsave(&sport->mod_sqp_lock, flags);
 376                         suppress_event = !sport->ibqp_sqp[IB_QPT_GSI];
 377                         if (sport->ibqp_sqp[IB_QPT_SMI])
 378                                 ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_SMI]);
 379                         if (!suppress_event)
 380                                 ehca_recover_sqp(sport->ibqp_sqp[IB_QPT_GSI]);
 381                         spin_unlock_irqrestore(&sport->mod_sqp_lock, flags);
 382
 383                         /* AQP1 was destroyed, ignore this event */
 384                         if (suppress_event)
 385                                 break;
 386
 387                         sport->port_state = IB_PORT_ACTIVE;
 388                         dispatch_port_event(shca, port, IB_EVENT_PORT_ACTIVE,
 389                                             "is active");
 390                         ehca_query_sma_attr(shca, port,
 391                                             &sport->saved_attr);
 392                 } else {
 393                         sport->port_state = IB_PORT_DOWN;
 394                         dispatch_port_event(shca, port, IB_EVENT_PORT_ERR,
 395                                             "is inactive");
 396                 }
 397                 break;
 398         case 0x31:
 399                 /* port configuration change
 400                  * disruptive change is caused by
 401                  * LID, PKEY or SM change
 402                  */
 403                 if (EHCA_BMASK_GET(NEQE_DISRUPTIVE, eqe)) {
 404                         ehca_warn(&shca->ib_device, "disruptive port "
 405                                   "%d configuration change", port);
 406
 407                         sport->port_state = IB_PORT_DOWN;
 408                         dispatch_port_event(shca, port, IB_EVENT_PORT_ERR,
 409                                             "is inactive");
 410
 411                         sport->port_state = IB_PORT_ACTIVE;
 412                         dispatch_port_event(shca, port, IB_EVENT_PORT_ACTIVE,
 413                                             "is active");
 414                         ehca_query_sma_attr(shca, port,
 415                                             &sport->saved_attr);
 416                 } else
 417                         notify_port_conf_change(shca, port);
 418                 break;
 419         case 0x32: /* adapter malfunction */
 420                 ehca_err(&shca->ib_device, "Adapter malfunction.");
 421                 break;
 422         case 0x33:  /* trace stopped */
 423                 ehca_err(&shca->ib_device, "Traced stopped.");
 424                 break;
 425         case 0x34: /* util async event */
 426                 spec_event = EHCA_BMASK_GET(NEQE_SPECIFIC_EVENT, eqe);
 427                 if (spec_event == 0x80) /* client reregister required */
 428                         dispatch_port_event(shca, port,
 429                                             IB_EVENT_CLIENT_REREGISTER,
 430                                             "client reregister req.");
 431                 else
 432                         ehca_warn(&shca->ib_device, "Unknown util async "
 433                                   "event %x on port %x", spec_event, port);
 434                 break;
 435         default:
 436                 ehca_err(&shca->ib_device, "Unknown event code: %x on %s.",
 437                          ec, shca->ib_device.name);
 438                 break;
 439         }
 440
 441         return;
 442 }
 443
 444 static inline void reset_eq_pending(struct ehca_cq *cq)
 445 {
 446         u64 CQx_EP;
 447         struct h_galpa gal = cq->galpas.kernel;
 448
 449         hipz_galpa_store_cq(gal, cqx_ep, 0x0);
 450         CQx_EP = hipz_galpa_load(gal, CQTEMM_OFFSET(cqx_ep));
 451
 452         return;
 453 }
 454
 455 irqreturn_t ehca_interrupt_neq(int irq, void *dev_id)
 456 {
 457         struct ehca_shca *shca = (struct ehca_shca*)dev_id;
 458
 459         tasklet_hi_schedule(&shca->neq.interrupt_task);
 460
 461         return IRQ_HANDLED;
 462 }
 463
 464 void ehca_tasklet_neq(unsigned long data)
 465 {
 466         struct ehca_shca *shca = (struct ehca_shca*)data;
 467         struct ehca_eqe *eqe;
 468         u64 ret;
 469
 470         eqe = (struct ehca_eqe *)ehca_poll_eq(shca, &shca->neq);
 471
 472         while (eqe) {
 473                 if (!EHCA_BMASK_GET(NEQE_COMPLETION_EVENT, eqe->entry))
 474                         parse_ec(shca, eqe->entry);
 475
 476                 eqe = (struct ehca_eqe *)ehca_poll_eq(shca, &shca->neq);
 477         }
 478
 479         ret = hipz_h_reset_event(shca->ipz_hca_handle,
 480                                  shca->neq.ipz_eq_handle, 0xFFFFFFFFFFFFFFFFL);
 481
 482         if (ret != H_SUCCESS)
 483                 ehca_err(&shca->ib_device, "Can't clear notification events.");
 484
 485         return;
 486 }
 487
 488 irqreturn_t ehca_interrupt_eq(int irq, void *dev_id)
 489 {
 490         struct ehca_shca *shca = (struct ehca_shca*)dev_id;
 491
 492         tasklet_hi_schedule(&shca->eq.interrupt_task);
 493
 494         return IRQ_HANDLED;
 495 }
 496
 497
 498 static inline void process_eqe(struct ehca_shca *shca, struct ehca_eqe *eqe)
 499 {
 500         u64 eqe_value;
 501         u32 token;
 502         struct ehca_cq *cq;
 503
 504         eqe_value = eqe->entry;
 505         ehca_dbg(&shca->ib_device, "eqe_value=%lx", eqe_value);
 506         if (EHCA_BMASK_GET(EQE_COMPLETION_EVENT, eqe_value)) {
 507                 ehca_dbg(&shca->ib_device, "Got completion event");
 508                 token = EHCA_BMASK_GET(EQE_CQ_TOKEN, eqe_value);
 509                 read_lock(&ehca_cq_idr_lock);
 510                 cq = idr_find(&ehca_cq_idr, token);
 511                 if (cq)
 512                         atomic_inc(&cq->nr_events);
 513                 read_unlock(&ehca_cq_idr_lock);
 514                 if (cq == NULL) {
 515                         ehca_err(&shca->ib_device,
 516                                  "Invalid eqe for non-existing cq token=%x",
 517                                  token);
 518                         return;
 519                 }
 520                 reset_eq_pending(cq);
 521                 if (ehca_scaling_code)
 522                         queue_comp_task(cq);
 523                 else {
 524                         comp_event_callback(cq);
 525                         if (atomic_dec_and_test(&cq->nr_events))
 526                                 wake_up(&cq->wait_completion);
 527                 }
 528         } else {
 529                 ehca_dbg(&shca->ib_device, "Got non completion event");
 530                 parse_identifier(shca, eqe_value);
 531         }
 532 }
 533
 534 void ehca_process_eq(struct ehca_shca *shca, int is_irq)
 535 {
 536         struct ehca_eq *eq = &shca->eq;
 537         struct ehca_eqe_cache_entry *eqe_cache = eq->eqe_cache;
 538         u64 eqe_value, ret;
 539         unsigned long flags;
 540         int eqe_cnt, i;
 541         int eq_empty = 0;
 542
 543         spin_lock_irqsave(&eq->irq_spinlock, flags);
 544         if (is_irq) {
 545                 const int max_query_cnt = 100;
 546                 int query_cnt = 0;
 547                 int int_state = 1;
 548                 do {
 549                         int_state = hipz_h_query_int_state(
 550                                 shca->ipz_hca_handle, eq->ist);
 551                         query_cnt++;
 552                         iosync();
 553                 } while (int_state && query_cnt < max_query_cnt);
 554                 if (unlikely((query_cnt == max_query_cnt)))
 555                         ehca_dbg(&shca->ib_device, "int_state=%x query_cnt=%x",
 556                                  int_state, query_cnt);
 557         }
 558
 559         /* read out all eqes */
 560         eqe_cnt = 0;
 561         do {
 562                 u32 token;
 563                 eqe_cache[eqe_cnt].eqe =
 564                         (struct ehca_eqe *)ehca_poll_eq(shca, eq);
 565                 if (!eqe_cache[eqe_cnt].eqe)
 566                         break;
 567                 eqe_value = eqe_cache[eqe_cnt].eqe->entry;
 568                 if (EHCA_BMASK_GET(EQE_COMPLETION_EVENT, eqe_value)) {
 569                         token = EHCA_BMASK_GET(EQE_CQ_TOKEN, eqe_value);
 570                         read_lock(&ehca_cq_idr_lock);
 571                         eqe_cache[eqe_cnt].cq = idr_find(&ehca_cq_idr, token);
 572                         if (eqe_cache[eqe_cnt].cq)
 573                                 atomic_inc(&eqe_cache[eqe_cnt].cq->nr_events);
 574                         read_unlock(&ehca_cq_idr_lock);
 575                         if (!eqe_cache[eqe_cnt].cq) {
 576                                 ehca_err(&shca->ib_device,
 577                                          "Invalid eqe for non-existing cq "
 578                                          "token=%x", token);
 579                                 continue;
 580                         }
 581                 } else
 582                         eqe_cache[eqe_cnt].cq = NULL;
 583                 eqe_cnt++;
 584         } while (eqe_cnt < EHCA_EQE_CACHE_SIZE);
 585         if (!eqe_cnt) {
 586                 if (is_irq)
 587                         ehca_dbg(&shca->ib_device,
 588                                  "No eqe found for irq event");
 589                 goto unlock_irq_spinlock;
 590         } else if (!is_irq) {
 591                 ret = hipz_h_eoi(eq->ist);
 592                 if (ret != H_SUCCESS)
 593                         ehca_err(&shca->ib_device,
 594                                  "bad return code EOI -rc = %ld\n", ret);
 595                 ehca_dbg(&shca->ib_device, "deadman found %x eqe", eqe_cnt);
 596         }
 597         if (unlikely(eqe_cnt == EHCA_EQE_CACHE_SIZE))
 598                 ehca_dbg(&shca->ib_device, "too many eqes for one irq event");
 599         /* enable irq for new packets */
 600         for (i = 0; i < eqe_cnt; i++) {
 601                 if (eq->eqe_cache[i].cq)
 602                         reset_eq_pending(eq->eqe_cache[i].cq);
 603         }
 604         /* check eq */
 605         spin_lock(&eq->spinlock);
 606         eq_empty = (!ipz_eqit_eq_peek_valid(&shca->eq.ipz_queue));
 607         spin_unlock(&eq->spinlock);
 608         /* call completion handler for cached eqes */
 609         for (i = 0; i < eqe_cnt; i++)
 610                 if (eq->eqe_cache[i].cq) {
 611                         if (ehca_scaling_code)
 612                                 queue_comp_task(eq->eqe_cache[i].cq);
 613                         else {
 614                                 struct ehca_cq *cq = eq->eqe_cache[i].cq;
 615                                 comp_event_callback(cq);
 616                                 if (atomic_dec_and_test(&cq->nr_events))
 617                                         wake_up(&cq->wait_completion);
 618                         }
 619                 } else {
 620                         ehca_dbg(&shca->ib_device, "Got non completion event");
 621                         parse_identifier(shca, eq->eqe_cache[i].eqe->entry);
 622                 }
 623         /* poll eq if not empty */
 624         if (eq_empty)
 625                 goto unlock_irq_spinlock;
 626         do {
 627                 struct ehca_eqe *eqe;
 628                 eqe = (struct ehca_eqe *)ehca_poll_eq(shca, &shca->eq);
 629                 if (!eqe)
 630                         break;
 631                 process_eqe(shca, eqe);
 632         } while (1);
 633
 634 unlock_irq_spinlock:
 635         spin_unlock_irqrestore(&eq->irq_spinlock, flags);
 636 }
 637
 638 void ehca_tasklet_eq(unsigned long data)
 639 {
 640         ehca_process_eq((struct ehca_shca*)data, 1);
 641 }
 642
 643 static inline int find_next_online_cpu(struct ehca_comp_pool *pool)
 644 {
 645         int cpu;
 646         unsigned long flags;
 647
 648         WARN_ON_ONCE(!in_interrupt());
 649         if (ehca_debug_level >= 3)
 650                 ehca_dmp(&cpu_online_map, sizeof(cpumask_t), "");
 651
 652         spin_lock_irqsave(&pool->last_cpu_lock, flags);
 653         cpu = next_cpu(pool->last_cpu, cpu_online_map);
 654         if (cpu == NR_CPUS)
 655                 cpu = first_cpu(cpu_online_map);
 656         pool->last_cpu = cpu;
 657         spin_unlock_irqrestore(&pool->last_cpu_lock, flags);
 658
 659         return cpu;
 660 }
 661
 662 static void __queue_comp_task(struct ehca_cq *__cq,
 663                               struct ehca_cpu_comp_task *cct)
 664 {
 665         unsigned long flags;
 666
 667         spin_lock_irqsave(&cct->task_lock, flags);
 668         spin_lock(&__cq->task_lock);
 669
 670         if (__cq->nr_callbacks == 0) {
 671                 __cq->nr_callbacks++;
 672                 list_add_tail(&__cq->entry, &cct->cq_list);
 673                 cct->cq_jobs++;
 674                 wake_up(&cct->wait_queue);
 675         } else
 676                 __cq->nr_callbacks++;
 677
 678         spin_unlock(&__cq->task_lock);
 679         spin_unlock_irqrestore(&cct->task_lock, flags);
 680 }
 681
 682 static void queue_comp_task(struct ehca_cq *__cq)
 683 {
 684         int cpu_id;
 685         struct ehca_cpu_comp_task *cct;
 686         int cq_jobs;
 687         unsigned long flags;
 688
 689         cpu_id = find_next_online_cpu(pool);
 690         BUG_ON(!cpu_online(cpu_id));
 691
 692         cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu_id);
 693         BUG_ON(!cct);
 694
 695         spin_lock_irqsave(&cct->task_lock, flags);
 696         cq_jobs = cct->cq_jobs;
 697         spin_unlock_irqrestore(&cct->task_lock, flags);
 698         if (cq_jobs > 0) {
 699                 cpu_id = find_next_online_cpu(pool);
 700                 cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu_id);
 701                 BUG_ON(!cct);
 702         }
 703
 704         __queue_comp_task(__cq, cct);
 705 }
 706
 707 static void run_comp_task(struct ehca_cpu_comp_task *cct)
 708 {
 709         struct ehca_cq *cq;
 710         unsigned long flags;
 711
 712         spin_lock_irqsave(&cct->task_lock, flags);
 713
 714         while (!list_empty(&cct->cq_list)) {
 715                 cq = list_entry(cct->cq_list.next, struct ehca_cq, entry);
 716                 spin_unlock_irqrestore(&cct->task_lock, flags);
 717
 718                 comp_event_callback(cq);
 719                 if (atomic_dec_and_test(&cq->nr_events))
 720                         wake_up(&cq->wait_completion);
 721
 722                 spin_lock_irqsave(&cct->task_lock, flags);
 723                 spin_lock(&cq->task_lock);
 724                 cq->nr_callbacks--;
 725                 if (!cq->nr_callbacks) {
 726                         list_del_init(cct->cq_list.next);
 727                         cct->cq_jobs--;
 728                 }
 729                 spin_unlock(&cq->task_lock);
 730         }
 731
 732         spin_unlock_irqrestore(&cct->task_lock, flags);
 733 }
 734
 735 static int comp_task(void *__cct)
 736 {
 737         struct ehca_cpu_comp_task *cct = __cct;
 738         int cql_empty;
 739         DECLARE_WAITQUEUE(wait, current);
 740
 741         set_current_state(TASK_INTERRUPTIBLE);
 742         while (!kthread_should_stop()) {
 743                 add_wait_queue(&cct->wait_queue, &wait);
 744
 745                 spin_lock_irq(&cct->task_lock);
 746                 cql_empty = list_empty(&cct->cq_list);
 747                 spin_unlock_irq(&cct->task_lock);
 748                 if (cql_empty)
 749                         schedule();
 750                 else
 751                         __set_current_state(TASK_RUNNING);
 752
 753                 remove_wait_queue(&cct->wait_queue, &wait);
 754
 755                 spin_lock_irq(&cct->task_lock);
 756                 cql_empty = list_empty(&cct->cq_list);
 757                 spin_unlock_irq(&cct->task_lock);
 758                 if (!cql_empty)
 759                         run_comp_task(__cct);
 760
 761                 set_current_state(TASK_INTERRUPTIBLE);
 762         }
 763         __set_current_state(TASK_RUNNING);
 764
 765         return 0;
 766 }
 767
 768 static struct task_struct *create_comp_task(struct ehca_comp_pool *pool,
 769                                             int cpu)
 770 {
 771         struct ehca_cpu_comp_task *cct;
 772
 773         cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
 774         spin_lock_init(&cct->task_lock);
 775         INIT_LIST_HEAD(&cct->cq_list);
 776         init_waitqueue_head(&cct->wait_queue);
 777         cct->task = kthread_create(comp_task, cct, "ehca_comp/%d", cpu);
 778
 779         return cct->task;
 780 }
 781
 782 static void destroy_comp_task(struct ehca_comp_pool *pool,
 783                               int cpu)
 784 {
 785         struct ehca_cpu_comp_task *cct;
 786         struct task_struct *task;
 787         unsigned long flags_cct;
 788
 789         cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
 790
 791         spin_lock_irqsave(&cct->task_lock, flags_cct);
 792
 793         task = cct->task;
 794         cct->task = NULL;
 795         cct->cq_jobs = 0;
 796
 797         spin_unlock_irqrestore(&cct->task_lock, flags_cct);
 798
 799         if (task)
 800                 kthread_stop(task);
 801 }
 802
 803 static void __cpuinit take_over_work(struct ehca_comp_pool *pool, int cpu)
 804 {
 805         struct ehca_cpu_comp_task *cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
 806         LIST_HEAD(list);
 807         struct ehca_cq *cq;
 808         unsigned long flags_cct;
 809
 810         spin_lock_irqsave(&cct->task_lock, flags_cct);
 811
 812         list_splice_init(&cct->cq_list, &list);
 813
 814         while (!list_empty(&list)) {
 815                 cq = list_entry(cct->cq_list.next, struct ehca_cq, entry);
 816
 817                 list_del(&cq->entry);
 818                 __queue_comp_task(cq, per_cpu_ptr(pool->cpu_comp_tasks,
 819                                                   smp_processor_id()));
 820         }
 821
 822         spin_unlock_irqrestore(&cct->task_lock, flags_cct);
 823
 824 }
 825
 826 static int __cpuinit comp_pool_callback(struct notifier_block *nfb,
 827                                         unsigned long action,
 828                                         void *hcpu)
 829 {
 830         unsigned int cpu = (unsigned long)hcpu;
 831         struct ehca_cpu_comp_task *cct;
 832
 833         switch (action) {
 834         case CPU_UP_PREPARE:
 835         case CPU_UP_PREPARE_FROZEN:
 836                 ehca_gen_dbg("CPU: %x (CPU_PREPARE)", cpu);
 837                 if (!create_comp_task(pool, cpu)) {
 838                         ehca_gen_err("Can't create comp_task for cpu: %x", cpu);
 839                         return NOTIFY_BAD;
 840                 }
 841                 break;
 842         case CPU_UP_CANCELED:
 843         case CPU_UP_CANCELED_FROZEN:
 844                 ehca_gen_dbg("CPU: %x (CPU_CANCELED)", cpu);
 845                 cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
 846                 kthread_bind(cct->task, any_online_cpu(cpu_online_map));
 847                 destroy_comp_task(pool, cpu);
 848                 break;
 849         case CPU_ONLINE:
 850         case CPU_ONLINE_FROZEN:
 851                 ehca_gen_dbg("CPU: %x (CPU_ONLINE)", cpu);
 852                 cct = per_cpu_ptr(pool->cpu_comp_tasks, cpu);
 853                 kthread_bind(cct->task, cpu);
 854                 wake_up_process(cct->task);
 855                 break;
 856         case CPU_DOWN_PREPARE:
 857         case CPU_DOWN_PREPARE_FROZEN:
 858                 ehca_gen_dbg("CPU: %x (CPU_DOWN_PREPARE)", cpu);
 859                 break;
 860         case CPU_DOWN_FAILED:
 861         case CPU_DOWN_FAILED_FROZEN:
 862                 ehca_gen_dbg("CPU: %x (CPU_DOWN_FAILED)", cpu);
 863                 break;
 864         case CPU_DEAD:
 865         case CPU_DEAD_FROZEN:
 866                 ehca_gen_dbg("CPU: %x (CPU_DEAD)", cpu);
 867                 destroy_comp_task(pool, cpu);
 868                 take_over_work(pool, cpu);
 869                 break;
 870         }
 871
 872         return NOTIFY_OK;
 873 }
 874
 875 static struct notifier_block comp_pool_callback_nb __cpuinitdata = {
 876         .notifier_call  = comp_pool_callback,
 877         .priority       = 0,
 878 };
 879
 880 int ehca_create_comp_pool(void)
 881 {
 882         int cpu;
 883         struct task_struct *task;
 884
 885         if (!ehca_scaling_code)
 886                 return 0;
 887
 888         pool = kzalloc(sizeof(struct ehca_comp_pool), GFP_KERNEL);
 889         if (pool == NULL)
 890                 return -ENOMEM;
 891
 892         spin_lock_init(&pool->last_cpu_lock);
 893         pool->last_cpu = any_online_cpu(cpu_online_map);
 894
 895         pool->cpu_comp_tasks = alloc_percpu(struct ehca_cpu_comp_task);
 896         if (pool->cpu_comp_tasks == NULL) {
 897                 kfree(pool);
 898                 return -EINVAL;
 899         }
 900
 901         for_each_online_cpu(cpu) {
 902                 task = create_comp_task(pool, cpu);
 903                 if (task) {
 904                         kthread_bind(task, cpu);
 905                         wake_up_process(task);
 906                 }
 907         }
 908
 909         register_hotcpu_notifier(&comp_pool_callback_nb);
 910
 911         printk(KERN_INFO "eHCA scaling code enabled\n");
 912
 913         return 0;
 914 }
 915
 916 void ehca_destroy_comp_pool(void)
 917 {
 918         int i;
 919
 920         if (!ehca_scaling_code)
 921                 return;
 922
 923         unregister_hotcpu_notifier(&comp_pool_callback_nb);
 924
 925         for (i = 0; i < NR_CPUS; i++) {
 926                 if (cpu_online(i))
 927                         destroy_comp_task(pool, i);
 928         }
 929         free_percpu(pool->cpu_comp_tasks);
 930         kfree(pool);
 931 }