pcie.c 22.1 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
/*
 * pcie.c
 *
 * Copyright (c) 2010 Isaku Yamahata <yamahata at valinux co jp>
 *                    VA Linux Systems Japan K.K.
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License along
 * with this program; if not, see <http://www.gnu.org/licenses/>.
 */

21
#include "qemu-common.h"
22 23 24 25
#include "hw/pci/pci_bridge.h"
#include "hw/pci/pcie.h"
#include "hw/pci/msix.h"
#include "hw/pci/msi.h"
26
#include "hw/pci/pci_bus.h"
27
#include "hw/pci/pcie_regs.h"
28
#include "qemu/range.h"
29
#include "qapi/qmp/qerror.h"
30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90

//#define DEBUG_PCIE
#ifdef DEBUG_PCIE
# define PCIE_DPRINTF(fmt, ...)                                         \
    fprintf(stderr, "%s:%d " fmt, __func__, __LINE__, ## __VA_ARGS__)
#else
# define PCIE_DPRINTF(fmt, ...) do {} while (0)
#endif
#define PCIE_DEV_PRINTF(dev, fmt, ...)                                  \
    PCIE_DPRINTF("%s:%x "fmt, (dev)->name, (dev)->devfn, ## __VA_ARGS__)


/***************************************************************************
 * pci express capability helper functions
 */
int pcie_cap_init(PCIDevice *dev, uint8_t offset, uint8_t type, uint8_t port)
{
    int pos;
    uint8_t *exp_cap;

    assert(pci_is_express(dev));

    pos = pci_add_capability(dev, PCI_CAP_ID_EXP, offset,
                                 PCI_EXP_VER2_SIZEOF);
    if (pos < 0) {
        return pos;
    }
    dev->exp.exp_cap = pos;
    exp_cap = dev->config + pos;

    /* capability register
       interrupt message number defaults to 0 */
    pci_set_word(exp_cap + PCI_EXP_FLAGS,
                 ((type << PCI_EXP_FLAGS_TYPE_SHIFT) & PCI_EXP_FLAGS_TYPE) |
                 PCI_EXP_FLAGS_VER2);

    /* device capability register
     * table 7-12:
     * roll based error reporting bit must be set by all
     * Functions conforming to the ECN, PCI Express Base
     * Specification, Revision 1.1., or subsequent PCI Express Base
     * Specification revisions.
     */
    pci_set_long(exp_cap + PCI_EXP_DEVCAP, PCI_EXP_DEVCAP_RBER);

    pci_set_long(exp_cap + PCI_EXP_LNKCAP,
                 (port << PCI_EXP_LNKCAP_PN_SHIFT) |
                 PCI_EXP_LNKCAP_ASPMS_0S |
                 PCI_EXP_LNK_MLW_1 |
                 PCI_EXP_LNK_LS_25);

    pci_set_word(exp_cap + PCI_EXP_LNKSTA,
                 PCI_EXP_LNK_MLW_1 | PCI_EXP_LNK_LS_25);

    pci_set_long(exp_cap + PCI_EXP_DEVCAP2,
                 PCI_EXP_DEVCAP2_EFF | PCI_EXP_DEVCAP2_EETLPP);

    pci_set_word(dev->wmask + pos, PCI_EXP_DEVCTL2_EETLPPB);
    return pos;
}

91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106
int pcie_endpoint_cap_init(PCIDevice *dev, uint8_t offset)
{
    uint8_t type = PCI_EXP_TYPE_ENDPOINT;

    /*
     * Windows guests will report Code 10, device cannot start, if
     * a regular Endpoint type is exposed on a root complex.  These
     * should instead be Root Complex Integrated Endpoints.
     */
    if (pci_bus_is_express(dev->bus) && pci_bus_is_root(dev->bus)) {
        type = PCI_EXP_TYPE_RC_END;
    }

    return pcie_cap_init(dev, offset, type, 0);
}

107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158
void pcie_cap_exit(PCIDevice *dev)
{
    pci_del_capability(dev, PCI_CAP_ID_EXP, PCI_EXP_VER2_SIZEOF);
}

uint8_t pcie_cap_get_type(const PCIDevice *dev)
{
    uint32_t pos = dev->exp.exp_cap;
    assert(pos > 0);
    return (pci_get_word(dev->config + pos + PCI_EXP_FLAGS) &
            PCI_EXP_FLAGS_TYPE) >> PCI_EXP_FLAGS_TYPE_SHIFT;
}

/* MSI/MSI-X */
/* pci express interrupt message number */
/* 7.8.2 PCI Express Capabilities Register: Interrupt Message Number */
void pcie_cap_flags_set_vector(PCIDevice *dev, uint8_t vector)
{
    uint8_t *exp_cap = dev->config + dev->exp.exp_cap;
    assert(vector < 32);
    pci_word_test_and_clear_mask(exp_cap + PCI_EXP_FLAGS, PCI_EXP_FLAGS_IRQ);
    pci_word_test_and_set_mask(exp_cap + PCI_EXP_FLAGS,
                               vector << PCI_EXP_FLAGS_IRQ_SHIFT);
}

uint8_t pcie_cap_flags_get_vector(PCIDevice *dev)
{
    return (pci_get_word(dev->config + dev->exp.exp_cap + PCI_EXP_FLAGS) &
            PCI_EXP_FLAGS_IRQ) >> PCI_EXP_FLAGS_IRQ_SHIFT;
}

void pcie_cap_deverr_init(PCIDevice *dev)
{
    uint32_t pos = dev->exp.exp_cap;
    pci_long_test_and_set_mask(dev->config + pos + PCI_EXP_DEVCAP,
                               PCI_EXP_DEVCAP_RBER);
    pci_long_test_and_set_mask(dev->wmask + pos + PCI_EXP_DEVCTL,
                               PCI_EXP_DEVCTL_CERE | PCI_EXP_DEVCTL_NFERE |
                               PCI_EXP_DEVCTL_FERE | PCI_EXP_DEVCTL_URRE);
    pci_long_test_and_set_mask(dev->w1cmask + pos + PCI_EXP_DEVSTA,
                               PCI_EXP_DEVSTA_CED | PCI_EXP_DEVSTA_NFED |
                               PCI_EXP_DEVSTA_URD | PCI_EXP_DEVSTA_URD);
}

void pcie_cap_deverr_reset(PCIDevice *dev)
{
    uint8_t *devctl = dev->config + dev->exp.exp_cap + PCI_EXP_DEVCTL;
    pci_long_test_and_clear_mask(devctl,
                                 PCI_EXP_DEVCTL_CERE | PCI_EXP_DEVCTL_NFERE |
                                 PCI_EXP_DEVCTL_FERE | PCI_EXP_DEVCTL_URRE);
}

159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185
static void hotplug_event_update_event_status(PCIDevice *dev)
{
    uint32_t pos = dev->exp.exp_cap;
    uint8_t *exp_cap = dev->config + pos;
    uint16_t sltctl = pci_get_word(exp_cap + PCI_EXP_SLTCTL);
    uint16_t sltsta = pci_get_word(exp_cap + PCI_EXP_SLTSTA);

    dev->exp.hpev_notified = (sltctl & PCI_EXP_SLTCTL_HPIE) &&
        (sltsta & sltctl & PCI_EXP_HP_EV_SUPPORTED);
}

static void hotplug_event_notify(PCIDevice *dev)
{
    bool prev = dev->exp.hpev_notified;

    hotplug_event_update_event_status(dev);

    if (prev == dev->exp.hpev_notified) {
        return;
    }

    /* Note: the logic above does not take into account whether interrupts
     * are masked. The result is that interrupt will be sent when it is
     * subsequently unmasked. This appears to be legal: Section 6.7.3.4:
     * The Port may optionally send an MSI when there are hot-plug events that
     * occur while interrupt generation is disabled, and interrupt generation is
     * subsequently enabled. */
186 187 188 189 190
    if (msix_enabled(dev)) {
        msix_notify(dev, pcie_cap_flags_get_vector(dev));
    } else if (msi_enabled(dev)) {
        msi_notify(dev, pcie_cap_flags_get_vector(dev));
    } else {
191
        pci_set_irq(dev, dev->exp.hpev_notified);
192 193 194
    }
}

I
Isaku Yamahata 已提交
195 196 197 198
static void hotplug_event_clear(PCIDevice *dev)
{
    hotplug_event_update_event_status(dev);
    if (!msix_enabled(dev) && !msi_enabled(dev) && !dev->exp.hpev_notified) {
199
        pci_irq_deassert(dev);
I
Isaku Yamahata 已提交
200 201 202
    }
}

203
/*
204
 * A PCI Express Hot-Plug Event has occurred, so update slot status register
205 206 207 208 209 210 211
 * and notify OS of the event if necessary.
 *
 * 6.7.3 PCI Express Hot-Plug Events
 * 6.7.3.4 Software Notification of Hot-Plug Events
 */
static void pcie_cap_slot_event(PCIDevice *dev, PCIExpressHotPlugEvent event)
{
212 213 214
    /* Minor optimization: if nothing changed - no event is needed. */
    if (pci_word_test_and_set_mask(dev->config + dev->exp.exp_cap +
                                   PCI_EXP_SLTSTA, event)) {
215 216
        return;
    }
217
    hotplug_event_notify(dev);
218 219
}

220 221 222
static void pcie_cap_slot_hotplug_common(PCIDevice *hotplug_dev,
                                         DeviceState *dev,
                                         uint8_t **exp_cap, Error **errp)
223
{
224 225
    *exp_cap = hotplug_dev->config + hotplug_dev->exp.exp_cap;
    uint16_t sltsta = pci_get_word(*exp_cap + PCI_EXP_SLTSTA);
226

227
    PCIE_DEV_PRINTF(PCI_DEVICE(dev), "hotplug state: 0x%x\n", sltsta);
228 229 230 231
    if (sltsta & PCI_EXP_SLTSTA_EIS) {
        /* the slot is electromechanically locked.
         * This error is propagated up to qdev and then to HMP/QMP.
         */
232
        error_setg_errno(errp, -EBUSY, "slot is electromechanically locked");
233
    }
234 235 236 237 238 239
}

void pcie_cap_slot_hotplug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
                              Error **errp)
{
    uint8_t *exp_cap;
240
    PCIDevice *pci_dev = PCI_DEVICE(dev);
241 242

    pcie_cap_slot_hotplug_common(PCI_DEVICE(hotplug_dev), dev, &exp_cap, errp);
243

244 245 246 247
    /* Don't send event when device is enabled during qemu machine creation:
     * it is present on boot, no hotplug event is necessary. We do send an
     * event when the device is disabled later. */
    if (!dev->hotplugged) {
248 249
        pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTSTA,
                                   PCI_EXP_SLTSTA_PDS);
250
        return;
251
    }
252

253 254 255 256 257 258
    /* TODO: multifunction hot-plug.
     * Right now, only a device of function = 0 is allowed to be
     * hot plugged/unplugged.
     */
    assert(PCI_FUNC(pci_dev->devfn) == 0);

259 260
    pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTSTA,
                               PCI_EXP_SLTSTA_PDS);
261 262
    pcie_cap_slot_event(PCI_DEVICE(hotplug_dev),
                        PCI_EXP_HP_EV_PDC | PCI_EXP_HP_EV_ABP);
263 264 265 266 267 268 269 270 271
}

void pcie_cap_slot_hot_unplug_cb(HotplugHandler *hotplug_dev, DeviceState *dev,
                                 Error **errp)
{
    uint8_t *exp_cap;

    pcie_cap_slot_hotplug_common(PCI_DEVICE(hotplug_dev), dev, &exp_cap, errp);

272
    pcie_cap_slot_push_attention_button(PCI_DEVICE(hotplug_dev));
273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294
}

/* pci express slot for pci express root/downstream port
   PCI express capability slot registers */
void pcie_cap_slot_init(PCIDevice *dev, uint16_t slot)
{
    uint32_t pos = dev->exp.exp_cap;

    pci_word_test_and_set_mask(dev->config + pos + PCI_EXP_FLAGS,
                               PCI_EXP_FLAGS_SLOT);

    pci_long_test_and_clear_mask(dev->config + pos + PCI_EXP_SLTCAP,
                                 ~PCI_EXP_SLTCAP_PSN);
    pci_long_test_and_set_mask(dev->config + pos + PCI_EXP_SLTCAP,
                               (slot << PCI_EXP_SLTCAP_PSN_SHIFT) |
                               PCI_EXP_SLTCAP_EIP |
                               PCI_EXP_SLTCAP_HPS |
                               PCI_EXP_SLTCAP_HPC |
                               PCI_EXP_SLTCAP_PIP |
                               PCI_EXP_SLTCAP_AIP |
                               PCI_EXP_SLTCAP_ABP);

295 296 297 298 299 300 301 302 303
    if (dev->cap_present & QEMU_PCIE_SLTCAP_PCP) {
        pci_long_test_and_set_mask(dev->config + pos + PCI_EXP_SLTCAP,
                                   PCI_EXP_SLTCAP_PCP);
        pci_word_test_and_clear_mask(dev->config + pos + PCI_EXP_SLTCTL,
                                     PCI_EXP_SLTCTL_PCC);
        pci_word_test_and_set_mask(dev->wmask + pos + PCI_EXP_SLTCTL,
                                   PCI_EXP_SLTCTL_PCC);
    }

304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327
    pci_word_test_and_clear_mask(dev->config + pos + PCI_EXP_SLTCTL,
                                 PCI_EXP_SLTCTL_PIC |
                                 PCI_EXP_SLTCTL_AIC);
    pci_word_test_and_set_mask(dev->config + pos + PCI_EXP_SLTCTL,
                               PCI_EXP_SLTCTL_PIC_OFF |
                               PCI_EXP_SLTCTL_AIC_OFF);
    pci_word_test_and_set_mask(dev->wmask + pos + PCI_EXP_SLTCTL,
                               PCI_EXP_SLTCTL_PIC |
                               PCI_EXP_SLTCTL_AIC |
                               PCI_EXP_SLTCTL_HPIE |
                               PCI_EXP_SLTCTL_CCIE |
                               PCI_EXP_SLTCTL_PDCE |
                               PCI_EXP_SLTCTL_ABPE);
    /* Although reading PCI_EXP_SLTCTL_EIC returns always 0,
     * make the bit writable here in order to detect 1b is written.
     * pcie_cap_slot_write_config() test-and-clear the bit, so
     * this bit always returns 0 to the guest.
     */
    pci_word_test_and_set_mask(dev->wmask + pos + PCI_EXP_SLTCTL,
                               PCI_EXP_SLTCTL_EIC);

    pci_word_test_and_set_mask(dev->w1cmask + pos + PCI_EXP_SLTSTA,
                               PCI_EXP_HP_EV_SUPPORTED);

328 329
    dev->exp.hpev_notified = false;

330 331
    qbus_set_hotplug_handler(BUS(pci_bridge_get_sec_bus(PCI_BRIDGE(dev))),
                             DEVICE(dev), NULL);
332 333 334 335 336
}

void pcie_cap_slot_reset(PCIDevice *dev)
{
    uint8_t *exp_cap = dev->config + dev->exp.exp_cap;
337 338 339 340
    uint8_t port_type = pcie_cap_get_type(dev);

    assert(port_type == PCI_EXP_TYPE_DOWNSTREAM ||
           port_type == PCI_EXP_TYPE_ROOT_PORT);
341 342 343 344 345 346 347 348 349 350 351 352 353 354

    PCIE_DEV_PRINTF(dev, "reset\n");

    pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTCTL,
                                 PCI_EXP_SLTCTL_EIC |
                                 PCI_EXP_SLTCTL_PIC |
                                 PCI_EXP_SLTCTL_AIC |
                                 PCI_EXP_SLTCTL_HPIE |
                                 PCI_EXP_SLTCTL_CCIE |
                                 PCI_EXP_SLTCTL_PDCE |
                                 PCI_EXP_SLTCTL_ABPE);
    pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTCTL,
                               PCI_EXP_SLTCTL_AIC_OFF);

355 356
    if (dev->cap_present & QEMU_PCIE_SLTCAP_PCP) {
        /* Downstream ports enforce device number 0. */
M
Michael S. Tsirkin 已提交
357 358
        bool populated = pci_bridge_get_sec_bus(PCI_BRIDGE(dev))->devices[0];
        uint16_t pic;
359 360 361 362 363 364 365 366 367 368 369

        if (populated) {
            pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTCTL,
                                         PCI_EXP_SLTCTL_PCC);
        } else {
            pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTCTL,
                                       PCI_EXP_SLTCTL_PCC);
        }

        pic = populated ? PCI_EXP_SLTCTL_PIC_ON : PCI_EXP_SLTCTL_PIC_OFF;
        pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTCTL, pic);
M
Michael S. Tsirkin 已提交
370
    }
371

372 373 374 375 376 377
    pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTSTA,
                                 PCI_EXP_SLTSTA_EIS |/* on reset,
                                                        the lock is released */
                                 PCI_EXP_SLTSTA_CC |
                                 PCI_EXP_SLTSTA_PDC |
                                 PCI_EXP_SLTSTA_ABP);
378

M
Michael S. Tsirkin 已提交
379
    hotplug_event_update_event_status(dev);
380 381
}

382 383 384 385 386
static void pcie_unplug_device(PCIBus *bus, PCIDevice *dev, void *opaque)
{
    object_unparent(OBJECT(dev));
}

387
void pcie_cap_slot_write_config(PCIDevice *dev,
388
                                uint32_t addr, uint32_t val, int len)
389 390 391 392 393
{
    uint32_t pos = dev->exp.exp_cap;
    uint8_t *exp_cap = dev->config + pos;
    uint16_t sltsta = pci_get_word(exp_cap + PCI_EXP_SLTSTA);

I
Isaku Yamahata 已提交
394 395 396 397
    if (ranges_overlap(addr, len, pos + PCI_EXP_SLTSTA, 2)) {
        hotplug_event_clear(dev);
    }

M
Michael S. Tsirkin 已提交
398 399 400 401 402 403 404 405 406 407 408 409
    if (!ranges_overlap(addr, len, pos + PCI_EXP_SLTCTL, 2)) {
        return;
    }

    if (pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTCTL,
                                     PCI_EXP_SLTCTL_EIC)) {
        sltsta ^= PCI_EXP_SLTSTA_EIS; /* toggle PCI_EXP_SLTSTA_EIS bit */
        pci_set_word(exp_cap + PCI_EXP_SLTSTA, sltsta);
        PCIE_DEV_PRINTF(dev, "PCI_EXP_SLTCTL_EIC: "
                        "sltsta -> 0x%02"PRIx16"\n",
                        sltsta);
    }
410

411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426
    /*
     * If the slot is polulated, power indicator is off and power
     * controller is off, it is safe to detach the devices.
     */
    if ((sltsta & PCI_EXP_SLTSTA_PDS) && (val & PCI_EXP_SLTCTL_PCC) &&
        ((val & PCI_EXP_SLTCTL_PIC_OFF) == PCI_EXP_SLTCTL_PIC_OFF)) {
            PCIBus *sec_bus = pci_bridge_get_sec_bus(PCI_BRIDGE(dev));
            pci_for_each_device(sec_bus, pci_bus_num(sec_bus),
                                pcie_unplug_device, NULL);

            pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTSTA,
                                         PCI_EXP_SLTSTA_PDS);
            pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTSTA,
                                       PCI_EXP_SLTSTA_PDC);
    }

427
    hotplug_event_notify(dev);
M
Michael S. Tsirkin 已提交
428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445

    /* 
     * 6.7.3.2 Command Completed Events
     *
     * Software issues a command to a hot-plug capable Downstream Port by
     * issuing a write transaction that targets any portion of the Port’s Slot
     * Control register. A single write to the Slot Control register is
     * considered to be a single command, even if the write affects more than
     * one field in the Slot Control register. In response to this transaction,
     * the Port must carry out the requested actions and then set the
     * associated status field for the command completed event. */

    /* Real hardware might take a while to complete requested command because
     * physical movement would be involved like locking the electromechanical
     * lock.  However in our case, command is completed instantaneously above,
     * so send a command completion event right now.
     */
    pcie_cap_slot_event(dev, PCI_EXP_HP_EV_CCI);
446 447
}

448 449 450 451 452 453 454
int pcie_cap_slot_post_load(void *opaque, int version_id)
{
    PCIDevice *dev = opaque;
    hotplug_event_update_event_status(dev);
    return 0;
}

455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491
void pcie_cap_slot_push_attention_button(PCIDevice *dev)
{
    pcie_cap_slot_event(dev, PCI_EXP_HP_EV_ABP);
}

/* root control/capabilities/status. PME isn't emulated for now */
void pcie_cap_root_init(PCIDevice *dev)
{
    pci_set_word(dev->wmask + dev->exp.exp_cap + PCI_EXP_RTCTL,
                 PCI_EXP_RTCTL_SECEE | PCI_EXP_RTCTL_SENFEE |
                 PCI_EXP_RTCTL_SEFEE);
}

void pcie_cap_root_reset(PCIDevice *dev)
{
    pci_set_word(dev->config + dev->exp.exp_cap + PCI_EXP_RTCTL, 0);
}

/* function level reset(FLR) */
void pcie_cap_flr_init(PCIDevice *dev)
{
    pci_long_test_and_set_mask(dev->config + dev->exp.exp_cap + PCI_EXP_DEVCAP,
                               PCI_EXP_DEVCAP_FLR);

    /* Although reading BCR_FLR returns always 0,
     * the bit is made writable here in order to detect the 1b is written
     * pcie_cap_flr_write_config() test-and-clear the bit, so
     * this bit always returns 0 to the guest.
     */
    pci_word_test_and_set_mask(dev->wmask + dev->exp.exp_cap + PCI_EXP_DEVCTL,
                               PCI_EXP_DEVCTL_BCR_FLR);
}

void pcie_cap_flr_write_config(PCIDevice *dev,
                               uint32_t addr, uint32_t val, int len)
{
    uint8_t *devctl = dev->config + dev->exp.exp_cap + PCI_EXP_DEVCTL;
I
Isaku Yamahata 已提交
492 493 494 495 496
    if (pci_get_word(devctl) & PCI_EXP_DEVCTL_BCR_FLR) {
        /* Clear PCI_EXP_DEVCTL_BCR_FLR after invoking the reset handler
           so the handler can detect FLR by looking at this bit. */
        pci_device_reset(dev);
        pci_word_test_and_clear_mask(devctl, PCI_EXP_DEVCTL_BCR_FLR);
497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575
    }
}

/* Alternative Routing-ID Interpretation (ARI) */
/* ari forwarding support for down stream port */
void pcie_cap_ari_init(PCIDevice *dev)
{
    uint32_t pos = dev->exp.exp_cap;
    pci_long_test_and_set_mask(dev->config + pos + PCI_EXP_DEVCAP2,
                               PCI_EXP_DEVCAP2_ARI);
    pci_long_test_and_set_mask(dev->wmask + pos + PCI_EXP_DEVCTL2,
                               PCI_EXP_DEVCTL2_ARI);
}

void pcie_cap_ari_reset(PCIDevice *dev)
{
    uint8_t *devctl2 = dev->config + dev->exp.exp_cap + PCI_EXP_DEVCTL2;
    pci_long_test_and_clear_mask(devctl2, PCI_EXP_DEVCTL2_ARI);
}

bool pcie_cap_is_ari_enabled(const PCIDevice *dev)
{
    if (!pci_is_express(dev)) {
        return false;
    }
    if (!dev->exp.exp_cap) {
        return false;
    }

    return pci_get_long(dev->config + dev->exp.exp_cap + PCI_EXP_DEVCTL2) &
        PCI_EXP_DEVCTL2_ARI;
}

/**************************************************************************
 * pci express extended capability allocation functions
 * uint16_t ext_cap_id (16 bit)
 * uint8_t cap_ver (4 bit)
 * uint16_t cap_offset (12 bit)
 * uint16_t ext_cap_size
 */

static uint16_t pcie_find_capability_list(PCIDevice *dev, uint16_t cap_id,
                                          uint16_t *prev_p)
{
    uint16_t prev = 0;
    uint16_t next;
    uint32_t header = pci_get_long(dev->config + PCI_CONFIG_SPACE_SIZE);

    if (!header) {
        /* no extended capability */
        next = 0;
        goto out;
    }
    for (next = PCI_CONFIG_SPACE_SIZE; next;
         prev = next, next = PCI_EXT_CAP_NEXT(header)) {

        assert(next >= PCI_CONFIG_SPACE_SIZE);
        assert(next <= PCIE_CONFIG_SPACE_SIZE - 8);

        header = pci_get_long(dev->config + next);
        if (PCI_EXT_CAP_ID(header) == cap_id) {
            break;
        }
    }

out:
    if (prev_p) {
        *prev_p = prev;
    }
    return next;
}

uint16_t pcie_find_capability(PCIDevice *dev, uint16_t cap_id)
{
    return pcie_find_capability_list(dev, cap_id, NULL);
}

static void pcie_ext_cap_set_next(PCIDevice *dev, uint16_t pos, uint16_t next)
{
576
    uint32_t header = pci_get_long(dev->config + pos);
577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
    assert(!(next & (PCI_EXT_CAP_ALIGN - 1)));
    header = (header & ~PCI_EXT_CAP_NEXT_MASK) |
        ((next << PCI_EXT_CAP_NEXT_SHIFT) & PCI_EXT_CAP_NEXT_MASK);
    pci_set_long(dev->config + pos, header);
}

/*
 * caller must supply valid (offset, size) * such that the range shouldn't
 * overlap with other capability or other registers.
 * This function doesn't check it.
 */
void pcie_add_capability(PCIDevice *dev,
                         uint16_t cap_id, uint8_t cap_ver,
                         uint16_t offset, uint16_t size)
{
    uint32_t header;
    uint16_t next;

    assert(offset >= PCI_CONFIG_SPACE_SIZE);
    assert(offset < offset + size);
    assert(offset + size < PCIE_CONFIG_SPACE_SIZE);
    assert(size >= 8);
    assert(pci_is_express(dev));

    if (offset == PCI_CONFIG_SPACE_SIZE) {
        header = pci_get_long(dev->config + offset);
        next = PCI_EXT_CAP_NEXT(header);
    } else {
        uint16_t prev;

        /* 0 is reserved cap id. use internally to find the last capability
           in the linked list */
        next = pcie_find_capability_list(dev, 0, &prev);

        assert(prev >= PCI_CONFIG_SPACE_SIZE);
        assert(next == 0);
        pcie_ext_cap_set_next(dev, prev, offset);
    }
    pci_set_long(dev->config + offset, PCI_EXT_CAP(cap_id, cap_ver, next));

    /* Make capability read-only by default */
    memset(dev->wmask + offset, 0, size);
    memset(dev->w1cmask + offset, 0, size);
    /* Check capability by default */
    memset(dev->cmask + offset, 0xFF, size);
}

/**************************************************************************
 * pci express extended capability helper functions
 */

/* ARI */
void pcie_ari_init(PCIDevice *dev, uint16_t offset, uint16_t nextfn)
{
    pcie_add_capability(dev, PCI_EXT_CAP_ID_ARI, PCI_ARI_VER,
                        offset, PCI_ARI_SIZEOF);
633
    pci_set_long(dev->config + offset + PCI_ARI_CAP, (nextfn & 0xff) << 8);
634
}