drivers/xen/events/events_fifo.c

   1 /*
   2  * Xen event channels (FIFO-based ABI)
   3  *
   4  * Copyright (C) 2013 Citrix Systems R&D ltd.
   5  *
   6  * This source code is free software; you can redistribute it and/or
   7  * modify it under the terms of the GNU General Public License as
   8  * published by the Free Software Foundation; either version 2 of the
   9  * License, or (at your option) any later version.
  10  *
  11  * Or, when distributed separately from the Linux kernel or
  12  * incorporated into other software packages, subject to the following
  13  * license:
  14  *
  15  * Permission is hereby granted, free of charge, to any person obtaining a copy
  16  * of this source file (the "Software"), to deal in the Software without
  17  * restriction, including without limitation the rights to use, copy, modify,
  18  * merge, publish, distribute, sublicense, and/or sell copies of the Software,
  19  * and to permit persons to whom the Software is furnished to do so, subject to
  20  * the following conditions:
  21  *
  22  * The above copyright notice and this permission notice shall be included in
  23  * all copies or substantial portions of the Software.
  24  *
  25  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  26  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  27  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  28  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  29  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  30  * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
  31  * IN THE SOFTWARE.
  32  */
  33
  34 #define pr_fmt(fmt) "xen:" KBUILD_MODNAME ": " fmt
  35
  36 #include <linux/linkage.h>
  37 #include <linux/interrupt.h>
  38 #include <linux/irq.h>
  39 #include <linux/smp.h>
  40 #include <linux/percpu.h>
  41 #include <linux/cpu.h>
  42
  43 #include <asm/barrier.h>
  44 #include <asm/sync_bitops.h>
  45 #include <asm/xen/hypercall.h>
  46 #include <asm/xen/hypervisor.h>
  47
  48 #include <xen/xen.h>
  49 #include <xen/xen-ops.h>
  50 #include <xen/events.h>
  51 #include <xen/interface/xen.h>
  52 #include <xen/interface/event_channel.h>
  53 #include <xen/page.h>
  54
  55 #include "events_internal.h"
  56
  57 #define EVENT_WORDS_PER_PAGE (XEN_PAGE_SIZE / sizeof(event_word_t))
  58 #define MAX_EVENT_ARRAY_PAGES (EVTCHN_FIFO_NR_CHANNELS / EVENT_WORDS_PER_PAGE)
  59
  60 struct evtchn_fifo_queue {
  61         uint32_t head[EVTCHN_FIFO_MAX_QUEUES];
  62 };
  63
  64 static DEFINE_PER_CPU(struct evtchn_fifo_control_block *, cpu_control_block);
  65 static DEFINE_PER_CPU(struct evtchn_fifo_queue, cpu_queue);
  66 static event_word_t *event_array[MAX_EVENT_ARRAY_PAGES] __read_mostly;
  67 static unsigned event_array_pages __read_mostly;
  68
  69 /*
  70  * sync_set_bit() and friends must be unsigned long aligned.
  71  */
  72 #if BITS_PER_LONG > 32
  73
  74 #define BM(w) (unsigned long *)((unsigned long)w & ~0x7UL)
  75 #define EVTCHN_FIFO_BIT(b, w) \
  76     (((unsigned long)w & 0x4UL) ? (EVTCHN_FIFO_ ##b + 32) : EVTCHN_FIFO_ ##b)
  77
  78 #else
  79
  80 #define BM(w) ((unsigned long *)(w))
  81 #define EVTCHN_FIFO_BIT(b, w) EVTCHN_FIFO_ ##b
  82
  83 #endif
  84
  85 static inline event_word_t *event_word_from_port(unsigned port)
  86 {
  87         unsigned i = port / EVENT_WORDS_PER_PAGE;
  88
  89         return event_array[i] + port % EVENT_WORDS_PER_PAGE;
  90 }
  91
  92 static unsigned evtchn_fifo_max_channels(void)
  93 {
  94         return EVTCHN_FIFO_NR_CHANNELS;
  95 }
  96
  97 static unsigned evtchn_fifo_nr_channels(void)
  98 {
  99         return event_array_pages * EVENT_WORDS_PER_PAGE;
 100 }
 101
 102 static int init_control_block(int cpu,
 103                               struct evtchn_fifo_control_block *control_block)
 104 {
 105         struct evtchn_fifo_queue *q = &per_cpu(cpu_queue, cpu);
 106         struct evtchn_init_control init_control;
 107         unsigned int i;
 108
 109         /* Reset the control block and the local HEADs. */
 110         clear_page(control_block);
 111         for (i = 0; i < EVTCHN_FIFO_MAX_QUEUES; i++)
 112                 q->head[i] = 0;
 113
 114         init_control.control_gfn = virt_to_gfn(control_block);
 115         init_control.offset      = 0;
 116         init_control.vcpu        = cpu;
 117
 118         return HYPERVISOR_event_channel_op(EVTCHNOP_init_control, &init_control);
 119 }
 120
 121 static void free_unused_array_pages(void)
 122 {
 123         unsigned i;
 124
 125         for (i = event_array_pages; i < MAX_EVENT_ARRAY_PAGES; i++) {
 126                 if (!event_array[i])
 127                         break;
 128                 free_page((unsigned long)event_array[i]);
 129                 event_array[i] = NULL;
 130         }
 131 }
 132
 133 static void init_array_page(event_word_t *array_page)
 134 {
 135         unsigned i;
 136
 137         for (i = 0; i < EVENT_WORDS_PER_PAGE; i++)
 138                 array_page[i] = 1 << EVTCHN_FIFO_MASKED;
 139 }
 140
 141 static int evtchn_fifo_setup(struct irq_info *info)
 142 {
 143         unsigned port = info->evtchn;
 144         unsigned new_array_pages;
 145         int ret;
 146
 147         new_array_pages = port / EVENT_WORDS_PER_PAGE + 1;
 148
 149         if (new_array_pages > MAX_EVENT_ARRAY_PAGES)
 150                 return -EINVAL;
 151
 152         while (event_array_pages < new_array_pages) {
 153                 void *array_page;
 154                 struct evtchn_expand_array expand_array;
 155
 156                 /* Might already have a page if we've resumed. */
 157                 array_page = event_array[event_array_pages];
 158                 if (!array_page) {
 159                         array_page = (void *)__get_free_page(GFP_KERNEL);
 160                         if (array_page == NULL) {
 161                                 ret = -ENOMEM;
 162                                 goto error;
 163                         }
 164                         event_array[event_array_pages] = array_page;
 165                 }
 166
 167                 /* Mask all events in this page before adding it. */
 168                 init_array_page(array_page);
 169
 170                 expand_array.array_gfn = virt_to_gfn(array_page);
 171
 172                 ret = HYPERVISOR_event_channel_op(EVTCHNOP_expand_array, &expand_array);
 173                 if (ret < 0)
 174                         goto error;
 175
 176                 event_array_pages++;
 177         }
 178         return 0;
 179
 180   error:
 181         if (event_array_pages == 0)
 182                 panic("xen: unable to expand event array with initial page (%d)\n", ret);
 183         else
 184                 pr_err("unable to expand event array (%d)\n", ret);
 185         free_unused_array_pages();
 186         return ret;
 187 }
 188
 189 static void evtchn_fifo_bind_to_cpu(struct irq_info *info, unsigned cpu)
 190 {
 191         /* no-op */
 192 }
 193
 194 static void evtchn_fifo_clear_pending(unsigned port)
 195 {
 196         event_word_t *word = event_word_from_port(port);
 197         sync_clear_bit(EVTCHN_FIFO_BIT(PENDING, word), BM(word));
 198 }
 199
 200 static void evtchn_fifo_set_pending(unsigned port)
 201 {
 202         event_word_t *word = event_word_from_port(port);
 203         sync_set_bit(EVTCHN_FIFO_BIT(PENDING, word), BM(word));
 204 }
 205
 206 static bool evtchn_fifo_is_pending(unsigned port)
 207 {
 208         event_word_t *word = event_word_from_port(port);
 209         return sync_test_bit(EVTCHN_FIFO_BIT(PENDING, word), BM(word));
 210 }
 211
 212 static bool evtchn_fifo_test_and_set_mask(unsigned port)
 213 {
 214         event_word_t *word = event_word_from_port(port);
 215         return sync_test_and_set_bit(EVTCHN_FIFO_BIT(MASKED, word), BM(word));
 216 }
 217
 218 static void evtchn_fifo_mask(unsigned port)
 219 {
 220         event_word_t *word = event_word_from_port(port);
 221         sync_set_bit(EVTCHN_FIFO_BIT(MASKED, word), BM(word));
 222 }
 223
 224 static bool evtchn_fifo_is_masked(unsigned port)
 225 {
 226         event_word_t *word = event_word_from_port(port);
 227         return sync_test_bit(EVTCHN_FIFO_BIT(MASKED, word), BM(word));
 228 }
 229 /*
 230  * Clear MASKED, spinning if BUSY is set.
 231  */
 232 static void clear_masked(volatile event_word_t *word)
 233 {
 234         event_word_t new, old, w;
 235
 236         w = *word;
 237
 238         do {
 239                 old = w & ~(1 << EVTCHN_FIFO_BUSY);
 240                 new = old & ~(1 << EVTCHN_FIFO_MASKED);
 241                 w = sync_cmpxchg(word, old, new);
 242         } while (w != old);
 243 }
 244
 245 static void evtchn_fifo_unmask(unsigned port)
 246 {
 247         event_word_t *word = event_word_from_port(port);
 248
 249         BUG_ON(!irqs_disabled());
 250
 251         clear_masked(word);
 252         if (evtchn_fifo_is_pending(port)) {
 253                 struct evtchn_unmask unmask = { .port = port };
 254                 (void)HYPERVISOR_event_channel_op(EVTCHNOP_unmask, &unmask);
 255         }
 256 }
 257
 258 static uint32_t clear_linked(volatile event_word_t *word)
 259 {
 260         event_word_t new, old, w;
 261
 262         w = *word;
 263
 264         do {
 265                 old = w;
 266                 new = (w & ~((1 << EVTCHN_FIFO_LINKED)
 267                              | EVTCHN_FIFO_LINK_MASK));
 268         } while ((w = sync_cmpxchg(word, old, new)) != old);
 269
 270         return w & EVTCHN_FIFO_LINK_MASK;
 271 }
 272
 273 static void handle_irq_for_port(unsigned port)
 274 {
 275         int irq;
 276
 277         irq = get_evtchn_to_irq(port);
 278         if (irq != -1)
 279                 generic_handle_irq(irq);
 280 }
 281
 282 static void consume_one_event(unsigned cpu,
 283                               struct evtchn_fifo_control_block *control_block,
 284                               unsigned priority, unsigned long *ready,
 285                               bool drop)
 286 {
 287         struct evtchn_fifo_queue *q = &per_cpu(cpu_queue, cpu);
 288         uint32_t head;
 289         unsigned port;
 290         event_word_t *word;
 291
 292         head = q->head[priority];
 293
 294         /*
 295          * Reached the tail last time?  Read the new HEAD from the
 296          * control block.
 297          */
 298         if (head == 0) {
 299                 virt_rmb(); /* Ensure word is up-to-date before reading head. */
 300                 head = control_block->head[priority];
 301         }
 302
 303         port = head;
 304         word = event_word_from_port(port);
 305         head = clear_linked(word);
 306
 307         /*
 308          * If the link is non-zero, there are more events in the
 309          * queue, otherwise the queue is empty.
 310          *
 311          * If the queue is empty, clear this priority from our local
 312          * copy of the ready word.
 313          */
 314         if (head == 0)
 315                 clear_bit(priority, ready);
 316
 317         if (evtchn_fifo_is_pending(port) && !evtchn_fifo_is_masked(port)) {
 318                 if (unlikely(drop))
 319                         pr_warn("Dropping pending event for port %u\n", port);
 320                 else
 321                         handle_irq_for_port(port);
 322         }
 323
 324         q->head[priority] = head;
 325 }
 326
 327 static void __evtchn_fifo_handle_events(unsigned cpu, bool drop)
 328 {
 329         struct evtchn_fifo_control_block *control_block;
 330         unsigned long ready;
 331         unsigned q;
 332
 333         control_block = per_cpu(cpu_control_block, cpu);
 334
 335         ready = xchg(&control_block->ready, 0);
 336
 337         while (ready) {
 338                 q = find_first_bit(&ready, EVTCHN_FIFO_MAX_QUEUES);
 339                 consume_one_event(cpu, control_block, q, &ready, drop);
 340                 ready |= xchg(&control_block->ready, 0);
 341         }
 342 }
 343
 344 static void evtchn_fifo_handle_events(unsigned cpu)
 345 {
 346         __evtchn_fifo_handle_events(cpu, false);
 347 }
 348
 349 static void evtchn_fifo_resume(void)
 350 {
 351         unsigned cpu;
 352
 353         for_each_possible_cpu(cpu) {
 354                 void *control_block = per_cpu(cpu_control_block, cpu);
 355                 int ret;
 356
 357                 if (!control_block)
 358                         continue;
 359
 360                 /*
 361                  * If this CPU is offline, take the opportunity to
 362                  * free the control block while it is not being
 363                  * used.
 364                  */
 365                 if (!cpu_online(cpu)) {
 366                         free_page((unsigned long)control_block);
 367                         per_cpu(cpu_control_block, cpu) = NULL;
 368                         continue;
 369                 }
 370
 371                 ret = init_control_block(cpu, control_block);
 372                 if (ret < 0)
 373                         BUG();
 374         }
 375
 376         /*
 377          * The event array starts out as empty again and is extended
 378          * as normal when events are bound.  The existing pages will
 379          * be reused.
 380          */
 381         event_array_pages = 0;
 382 }
 383
 384 static const struct evtchn_ops evtchn_ops_fifo = {
 385         .max_channels      = evtchn_fifo_max_channels,
 386         .nr_channels       = evtchn_fifo_nr_channels,
 387         .setup             = evtchn_fifo_setup,
 388         .bind_to_cpu       = evtchn_fifo_bind_to_cpu,
 389         .clear_pending     = evtchn_fifo_clear_pending,
 390         .set_pending       = evtchn_fifo_set_pending,
 391         .is_pending        = evtchn_fifo_is_pending,
 392         .test_and_set_mask = evtchn_fifo_test_and_set_mask,
 393         .mask              = evtchn_fifo_mask,
 394         .unmask            = evtchn_fifo_unmask,
 395         .handle_events     = evtchn_fifo_handle_events,
 396         .resume            = evtchn_fifo_resume,
 397 };
 398
 399 static int evtchn_fifo_alloc_control_block(unsigned cpu)
 400 {
 401         void *control_block = NULL;
 402         int ret = -ENOMEM;
 403
 404         control_block = (void *)__get_free_page(GFP_KERNEL);
 405         if (control_block == NULL)
 406                 goto error;
 407
 408         ret = init_control_block(cpu, control_block);
 409         if (ret < 0)
 410                 goto error;
 411
 412         per_cpu(cpu_control_block, cpu) = control_block;
 413
 414         return 0;
 415
 416   error:
 417         free_page((unsigned long)control_block);
 418         return ret;
 419 }
 420
 421 static int evtchn_fifo_cpu_notification(struct notifier_block *self,
 422                                                   unsigned long action,
 423                                                   void *hcpu)
 424 {
 425         int cpu = (long)hcpu;
 426         int ret = 0;
 427
 428         switch (action) {
 429         case CPU_UP_PREPARE:
 430                 if (!per_cpu(cpu_control_block, cpu))
 431                         ret = evtchn_fifo_alloc_control_block(cpu);
 432                 break;
 433         case CPU_DEAD:
 434                 __evtchn_fifo_handle_events(cpu, true);
 435                 break;
 436         default:
 437                 break;
 438         }
 439         return ret < 0 ? NOTIFY_BAD : NOTIFY_OK;
 440 }
 441
 442 static struct notifier_block evtchn_fifo_cpu_notifier = {
 443         .notifier_call  = evtchn_fifo_cpu_notification,
 444 };
 445
 446 int __init xen_evtchn_fifo_init(void)
 447 {
 448         int cpu = get_cpu();
 449         int ret;
 450
 451         ret = evtchn_fifo_alloc_control_block(cpu);
 452         if (ret < 0)
 453                 goto out;
 454
 455         pr_info("Using FIFO-based ABI\n");
 456
 457         evtchn_ops = &evtchn_ops_fifo;
 458
 459         register_cpu_notifier(&evtchn_fifo_cpu_notifier);
 460 out:
 461         put_cpu();
 462         return ret;
 463 }