aerdrv_core.c 22.4 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
/*
 * drivers/pci/pcie/aer/aerdrv_core.c
 *
 * This file is subject to the terms and conditions of the GNU General Public
 * License.  See the file "COPYING" in the main directory of this archive
 * for more details.
 *
 * This file implements the core part of PCI-Express AER. When an pci-express
 * error is delivered, an error message will be collected and printed to
 * console, then, an error recovery procedure will be executed by following
 * the pci error recovery rules.
 *
 * Copyright (C) 2006 Intel Corp.
 *	Tom Long Nguyen (tom.l.nguyen@intel.com)
 *	Zhang Yanmin (yanmin.zhang@intel.com)
 *
 */

#include <linux/module.h>
#include <linux/pci.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/pm.h>
#include <linux/suspend.h>
#include <linux/delay.h>
26
#include <linux/slab.h>
27
#include <linux/kfifo.h>
28 29
#include "aerdrv.h"

30 31
static bool forceload;
static bool nosourceid;
32
module_param(forceload, bool, 0);
33
module_param(nosourceid, bool, 0);
34

35 36 37
#define	PCI_EXP_AER_FLAGS	(PCI_EXP_DEVCTL_CERE | PCI_EXP_DEVCTL_NFERE | \
				 PCI_EXP_DEVCTL_FERE | PCI_EXP_DEVCTL_URRE)

38 39
int pci_enable_pcie_error_reporting(struct pci_dev *dev)
{
40
	if (pcie_aer_get_firmware_first(dev))
41 42
		return -EIO;

43
	if (!pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR))
44 45
		return -EIO;

46
	return pcie_capability_set_word(dev, PCI_EXP_DEVCTL, PCI_EXP_AER_FLAGS);
47
}
48
EXPORT_SYMBOL_GPL(pci_enable_pcie_error_reporting);
49 50 51

int pci_disable_pcie_error_reporting(struct pci_dev *dev)
{
52
	if (pcie_aer_get_firmware_first(dev))
53 54
		return -EIO;

55 56
	return pcie_capability_clear_word(dev, PCI_EXP_DEVCTL,
					  PCI_EXP_AER_FLAGS);
57
}
58
EXPORT_SYMBOL_GPL(pci_disable_pcie_error_reporting);
59 60 61 62

int pci_cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
{
	int pos;
63
	u32 status;
64

65
	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
66 67 68 69
	if (!pos)
		return -EIO;

	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, &status);
70 71
	if (status)
		pci_write_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, status);
72 73 74

	return 0;
}
75
EXPORT_SYMBOL_GPL(pci_cleanup_aer_uncorrect_error_status);
76

77 78 79 80 81
/**
 * add_error_device - list device to be handled
 * @e_info: pointer to error info
 * @dev: pointer to pci_dev to be added
 */
Z
Zhang, Yanmin 已提交
82 83 84 85 86
static int add_error_device(struct aer_err_info *e_info, struct pci_dev *dev)
{
	if (e_info->error_dev_num < AER_MAX_MULTI_ERR_DEVICES) {
		e_info->dev[e_info->error_dev_num] = dev;
		e_info->error_dev_num++;
87
		return 0;
88
	}
89
	return -ENOSPC;
Z
Zhang, Yanmin 已提交
90 91
}

92 93
#define	PCI_BUS(x)	(((x) >> 8) & 0xff)

94 95 96 97 98 99
/**
 * is_error_source - check whether the device is source of reported error
 * @dev: pointer to pci_dev to be checked
 * @e_info: pointer to reported error info
 */
static bool is_error_source(struct pci_dev *dev, struct aer_err_info *e_info)
100 101
{
	int pos;
102
	u32 status, mask;
103 104 105 106 107 108 109
	u16 reg16;

	/*
	 * When bus id is equal to 0, it might be a bad id
	 * reported by root port.
	 */
	if (!nosourceid && (PCI_BUS(e_info->id) != 0)) {
110 111
		/* Device ID match? */
		if (e_info->id == ((dev->bus->number << 8) | dev->devfn))
112
			return true;
Z
Zhang, Yanmin 已提交
113

114
		/* Continue id comparing if there is no multiple error */
H
Hidetoshi Seto 已提交
115
		if (!e_info->multi_error_valid)
116
			return false;
117 118 119
	}

	/*
Z
Zhang, Yanmin 已提交
120 121 122 123 124
	 * When either
	 *      1) nosourceid==y;
	 *      2) bus id is equal to 0. Some ports might lose the bus
	 *              id of error source id;
	 *      3) There are multiple errors and prior id comparing fails;
125
	 * We check AER status registers to find possible reporter.
126 127
	 */
	if (atomic_read(&dev->enable_cnt) == 0)
128 129
		return false;

130
	/* Check if AER is enabled */
131 132
	pcie_capability_read_word(dev, PCI_EXP_DEVCTL, &reg16);
	if (!(reg16 & PCI_EXP_AER_FLAGS))
133
		return false;
134

135 136
	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
	if (!pos)
137
		return false;
138

139
	/* Check if error is recorded */
140
	if (e_info->severity == AER_CORRECTABLE) {
141 142
		pci_read_config_dword(dev, pos + PCI_ERR_COR_STATUS, &status);
		pci_read_config_dword(dev, pos + PCI_ERR_COR_MASK, &mask);
143
	} else {
144 145
		pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, &status);
		pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_MASK, &mask);
146
	}
147 148
	if (status & ~mask)
		return true;
149

150 151
	return false;
}
Z
Zhang, Yanmin 已提交
152

153 154 155 156 157
static int find_device_iter(struct pci_dev *dev, void *data)
{
	struct aer_err_info *e_info = (struct aer_err_info *)data;

	if (is_error_source(dev, e_info)) {
158 159 160 161 162 163
		/* List this device */
		if (add_error_device(e_info, dev)) {
			/* We cannot handle more... Stop iteration */
			/* TODO: Should print error message here? */
			return 1;
		}
164 165 166 167 168 169

		/* If there is only a single error, stop iteration */
		if (!e_info->multi_error_valid)
			return 1;
	}
	return 0;
170 171 172 173
}

/**
 * find_source_device - search through device hierarchy for source device
R
Randy Dunlap 已提交
174
 * @parent: pointer to Root Port pci_dev data structure
175
 * @e_info: including detailed error information such like id
176
 *
177 178 179
 * Return true if found.
 *
 * Invoked by DPC when error is detected at the Root Port.
180 181 182
 * Caller of this function must set id, severity, and multi_error_valid of
 * struct aer_err_info pointed by @e_info properly.  This function must fill
 * e_info->error_dev_num and e_info->dev[], based on the given information.
R
Randy Dunlap 已提交
183
 */
184
static bool find_source_device(struct pci_dev *parent,
185
		struct aer_err_info *e_info)
186 187
{
	struct pci_dev *dev = parent;
188
	int result;
189

190 191 192
	/* Must reset in this function */
	e_info->error_dev_num = 0;

193
	/* Is Root Port an agent that sends error message? */
194 195
	result = find_device_iter(dev, e_info);
	if (result)
196
		return true;
197

198
	pci_walk_bus(parent->subordinate, find_device_iter, e_info);
199 200 201 202 203 204 205 206

	if (!e_info->error_dev_num) {
		dev_printk(KERN_DEBUG, &parent->dev,
				"can't find device of ID%04x\n",
				e_info->id);
		return false;
	}
	return true;
207 208
}

209
static int report_error_detected(struct pci_dev *dev, void *data)
210 211
{
	pci_ers_result_t vote;
212
	const struct pci_error_handlers *err_handler;
213 214 215
	struct aer_broadcast_data *result_data;
	result_data = (struct aer_broadcast_data *) data;

216
	device_lock(&dev->dev);
217 218 219 220 221 222 223 224 225 226 227 228 229 230
	dev->error_state = result_data->state;

	if (!dev->driver ||
		!dev->driver->err_handler ||
		!dev->driver->err_handler->error_detected) {
		if (result_data->state == pci_channel_io_frozen &&
			!(dev->hdr_type & PCI_HEADER_TYPE_BRIDGE)) {
			/*
			 * In case of fatal recovery, if one of down-
			 * stream device has no driver. We might be
			 * unable to recover because a later insmod
			 * of a driver for this device is unaware of
			 * its hw state.
			 */
231 232 233
			dev_printk(KERN_DEBUG, &dev->dev, "device has %s\n",
				   dev->driver ?
				   "no AER-aware driver" : "no driver");
234
		}
235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252

		/*
		 * If there's any device in the subtree that does not
		 * have an error_detected callback, returning
		 * PCI_ERS_RESULT_NO_AER_DRIVER prevents calling of
		 * the subsequent mmio_enabled/slot_reset/resume
		 * callbacks of "any" device in the subtree. All the
		 * devices in the subtree are left in the error state
		 * without recovery.
		 */

		if (!(dev->hdr_type & PCI_HEADER_TYPE_BRIDGE))
			vote = PCI_ERS_RESULT_NO_AER_DRIVER;
		else
			vote = PCI_ERS_RESULT_NONE;
	} else {
		err_handler = dev->driver->err_handler;
		vote = err_handler->error_detected(dev, result_data->state);
253 254 255
	}

	result_data->result = merge_result(result_data->result, vote);
256
	device_unlock(&dev->dev);
257
	return 0;
258 259
}

260
static int report_mmio_enabled(struct pci_dev *dev, void *data)
261 262
{
	pci_ers_result_t vote;
263
	const struct pci_error_handlers *err_handler;
264 265 266
	struct aer_broadcast_data *result_data;
	result_data = (struct aer_broadcast_data *) data;

267
	device_lock(&dev->dev);
268 269 270
	if (!dev->driver ||
		!dev->driver->err_handler ||
		!dev->driver->err_handler->mmio_enabled)
271
		goto out;
272 273 274 275

	err_handler = dev->driver->err_handler;
	vote = err_handler->mmio_enabled(dev);
	result_data->result = merge_result(result_data->result, vote);
276 277
out:
	device_unlock(&dev->dev);
278
	return 0;
279 280
}

281
static int report_slot_reset(struct pci_dev *dev, void *data)
282 283
{
	pci_ers_result_t vote;
284
	const struct pci_error_handlers *err_handler;
285 286 287
	struct aer_broadcast_data *result_data;
	result_data = (struct aer_broadcast_data *) data;

288
	device_lock(&dev->dev);
289 290 291
	if (!dev->driver ||
		!dev->driver->err_handler ||
		!dev->driver->err_handler->slot_reset)
292
		goto out;
293 294 295 296

	err_handler = dev->driver->err_handler;
	vote = err_handler->slot_reset(dev);
	result_data->result = merge_result(result_data->result, vote);
297 298
out:
	device_unlock(&dev->dev);
299
	return 0;
300 301
}

302
static int report_resume(struct pci_dev *dev, void *data)
303
{
304
	const struct pci_error_handlers *err_handler;
305

306
	device_lock(&dev->dev);
307 308 309 310
	dev->error_state = pci_channel_io_normal;

	if (!dev->driver ||
		!dev->driver->err_handler ||
H
Hidetoshi Seto 已提交
311
		!dev->driver->err_handler->resume)
312
		goto out;
313 314 315

	err_handler = dev->driver->err_handler;
	err_handler->resume(dev);
316 317
out:
	device_unlock(&dev->dev);
318
	return 0;
319 320 321 322
}

/**
 * broadcast_error_message - handle message broadcast to downstream drivers
R
Randy Dunlap 已提交
323
 * @dev: pointer to from where in a hierarchy message is broadcasted down
324
 * @state: error state
R
Randy Dunlap 已提交
325 326
 * @error_mesg: message to print
 * @cb: callback to be broadcasted
327 328 329 330
 *
 * Invoked during error recovery process. Once being invoked, the content
 * of error severity will be broadcasted to all downstream drivers in a
 * hierarchy in question.
R
Randy Dunlap 已提交
331
 */
332 333 334
static pci_ers_result_t broadcast_error_message(struct pci_dev *dev,
	enum pci_channel_state state,
	char *error_mesg,
335
	int (*cb)(struct pci_dev *, void *))
336 337 338
{
	struct aer_broadcast_data result_data;

339
	dev_printk(KERN_DEBUG, &dev->dev, "broadcast %s message\n", error_mesg);
340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359
	result_data.state = state;
	if (cb == report_error_detected)
		result_data.result = PCI_ERS_RESULT_CAN_RECOVER;
	else
		result_data.result = PCI_ERS_RESULT_RECOVERED;

	if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
		/*
		 * If the error is reported by a bridge, we think this error
		 * is related to the downstream link of the bridge, so we
		 * do error recovery on all subordinates of the bridge instead
		 * of the bridge and clear the error status of the bridge.
		 */
		if (cb == report_error_detected)
			dev->error_state = state;
		pci_walk_bus(dev->subordinate, cb, &result_data);
		if (cb == report_resume) {
			pci_cleanup_aer_uncorrect_error_status(dev);
			dev->error_state = pci_channel_io_normal;
		}
360
	} else {
361 362 363 364 365 366 367 368 369 370
		/*
		 * If the error is reported by an end point, we think this
		 * error is related to the upstream link of the end point.
		 */
		pci_walk_bus(dev->bus, cb, &result_data);
	}

	return result_data.result;
}

371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417
/**
 * aer_do_secondary_bus_reset - perform secondary bus reset
 * @dev: pointer to bridge's pci_dev data structure
 *
 * Invoked when performing link reset at Root Port or Downstream Port.
 */
void aer_do_secondary_bus_reset(struct pci_dev *dev)
{
	u16 p2p_ctrl;

	/* Assert Secondary Bus Reset */
	pci_read_config_word(dev, PCI_BRIDGE_CONTROL, &p2p_ctrl);
	p2p_ctrl |= PCI_BRIDGE_CTL_BUS_RESET;
	pci_write_config_word(dev, PCI_BRIDGE_CONTROL, p2p_ctrl);

	/*
	 * we should send hot reset message for 2ms to allow it time to
	 * propagate to all downstream ports
	 */
	msleep(2);

	/* De-assert Secondary Bus Reset */
	p2p_ctrl &= ~PCI_BRIDGE_CTL_BUS_RESET;
	pci_write_config_word(dev, PCI_BRIDGE_CONTROL, p2p_ctrl);

	/*
	 * System software must wait for at least 100ms from the end
	 * of a reset of one or more device before it is permitted
	 * to issue Configuration Requests to those devices.
	 */
	msleep(200);
}

/**
 * default_downstream_reset_link - default reset function for Downstream Port
 * @dev: pointer to downstream port's pci_dev data structure
 *
 * Invoked when performing link reset at Downstream Port w/ no aer driver.
 */
static pci_ers_result_t default_downstream_reset_link(struct pci_dev *dev)
{
	aer_do_secondary_bus_reset(dev);
	dev_printk(KERN_DEBUG, &dev->dev,
		"Downstream Port link has been reset\n");
	return PCI_ERS_RESULT_RECOVERED;
}

418 419
static int find_aer_service_iter(struct device *device, void *data)
{
420
	struct pcie_port_service_driver *service_driver, **drv;
421

422
	drv = (struct pcie_port_service_driver **) data;
423

424 425 426
	if (device->bus == &pcie_port_bus_type && device->driver) {
		service_driver = to_service_driver(device->driver);
		if (service_driver->service == PCIE_PORT_SERVICE_AER) {
427
			*drv = service_driver;
428
			return 1;
429 430 431 432 433 434
		}
	}

	return 0;
}

435
static struct pcie_port_service_driver *find_aer_service(struct pci_dev *dev)
436
{
437 438 439 440 441
	struct pcie_port_service_driver *drv = NULL;

	device_for_each_child(&dev->dev, &drv, find_aer_service_iter);

	return drv;
442 443
}

444
static pci_ers_result_t reset_link(struct pci_dev *dev)
445 446 447
{
	struct pci_dev *udev;
	pci_ers_result_t status;
448
	struct pcie_port_service_driver *driver;
449

450 451
	if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
		/* Reset this port for all subordinates */
452
		udev = dev;
453 454
	} else {
		/* Reset the upstream component (likely downstream port) */
455
		udev = dev->bus->self;
456
	}
457

458 459
	/* Use the aer driver of the component firstly */
	driver = find_aer_service(udev);
460

461 462
	if (driver && driver->reset_link) {
		status = driver->reset_link(udev);
463
	} else if (pci_pcie_type(udev) == PCI_EXP_TYPE_DOWNSTREAM) {
464 465 466 467 468 469
		status = default_downstream_reset_link(udev);
	} else {
		dev_printk(KERN_DEBUG, &dev->dev,
			"no link-reset support at upstream device %s\n",
			pci_name(udev));
		return PCI_ERS_RESULT_DISCONNECT;
470 471 472
	}

	if (status != PCI_ERS_RESULT_RECOVERED) {
473 474 475
		dev_printk(KERN_DEBUG, &dev->dev,
			"link reset at upstream device %s failed\n",
			pci_name(udev));
476 477 478 479 480 481 482 483 484 485 486 487 488 489
		return PCI_ERS_RESULT_DISCONNECT;
	}

	return status;
}

/**
 * do_recovery - handle nonfatal/fatal error recovery process
 * @dev: pointer to a pci_dev data structure of agent detecting an error
 * @severity: error severity type
 *
 * Invoked when an error is nonfatal/fatal. Once being invoked, broadcast
 * error detected message to all downstream drivers within a hierarchy in
 * question and return the returned code.
R
Randy Dunlap 已提交
490
 */
491
static void do_recovery(struct pci_dev *dev, int severity)
492 493 494 495 496 497 498 499 500 501 502 503 504 505 506
{
	pci_ers_result_t status, result = PCI_ERS_RESULT_RECOVERED;
	enum pci_channel_state state;

	if (severity == AER_FATAL)
		state = pci_channel_io_frozen;
	else
		state = pci_channel_io_normal;

	status = broadcast_error_message(dev,
			state,
			"error_detected",
			report_error_detected);

	if (severity == AER_FATAL) {
507
		result = reset_link(dev);
H
Hidetoshi Seto 已提交
508 509
		if (result != PCI_ERS_RESULT_RECOVERED)
			goto failed;
510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529
	}

	if (status == PCI_ERS_RESULT_CAN_RECOVER)
		status = broadcast_error_message(dev,
				state,
				"mmio_enabled",
				report_mmio_enabled);

	if (status == PCI_ERS_RESULT_NEED_RESET) {
		/*
		 * TODO: Should call platform-specific
		 * functions to reset slot before calling
		 * drivers' slot_reset callbacks?
		 */
		status = broadcast_error_message(dev,
				state,
				"slot_reset",
				report_slot_reset);
	}

H
Hidetoshi Seto 已提交
530 531 532 533
	if (status != PCI_ERS_RESULT_RECOVERED)
		goto failed;

	broadcast_error_message(dev,
534 535 536 537
				state,
				"resume",
				report_resume);

538
	dev_info(&dev->dev, "AER: Device recovery successful\n");
H
Hidetoshi Seto 已提交
539 540 541 542
	return;

failed:
	/* TODO: Should kernel panic here? */
543
	dev_info(&dev->dev, "AER: Device recovery failed\n");
544 545 546 547 548 549 550 551 552
}

/**
 * handle_error_source - handle logging error into an event log
 * @aerdev: pointer to pcie_device data structure of the root port
 * @dev: pointer to pci_dev data structure of error source device
 * @info: comprehensive error information
 *
 * Invoked when an error being detected by Root Port.
R
Randy Dunlap 已提交
553
 */
554
static void handle_error_source(struct pcie_device *aerdev,
555
	struct pci_dev *dev,
556
	struct aer_err_info *info)
557 558 559
{
	int pos;

560
	if (info->severity == AER_CORRECTABLE) {
561 562 563 564
		/*
		 * Correctable error does not need software intevention.
		 * No need to go through error recovery process.
		 */
565
		pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
566 567
		if (pos)
			pci_write_config_dword(dev, pos + PCI_ERR_COR_STATUS,
568
					info->status);
H
Hidetoshi Seto 已提交
569
	} else
570
		do_recovery(dev, info->severity);
571 572
}

573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632
#ifdef CONFIG_ACPI_APEI_PCIEAER
static void aer_recover_work_func(struct work_struct *work);

#define AER_RECOVER_RING_ORDER		4
#define AER_RECOVER_RING_SIZE		(1 << AER_RECOVER_RING_ORDER)

struct aer_recover_entry
{
	u8	bus;
	u8	devfn;
	u16	domain;
	int	severity;
};

static DEFINE_KFIFO(aer_recover_ring, struct aer_recover_entry,
		    AER_RECOVER_RING_SIZE);
/*
 * Mutual exclusion for writers of aer_recover_ring, reader side don't
 * need lock, because there is only one reader and lock is not needed
 * between reader and writer.
 */
static DEFINE_SPINLOCK(aer_recover_ring_lock);
static DECLARE_WORK(aer_recover_work, aer_recover_work_func);

void aer_recover_queue(int domain, unsigned int bus, unsigned int devfn,
		       int severity)
{
	unsigned long flags;
	struct aer_recover_entry entry = {
		.bus		= bus,
		.devfn		= devfn,
		.domain		= domain,
		.severity	= severity,
	};

	spin_lock_irqsave(&aer_recover_ring_lock, flags);
	if (kfifo_put(&aer_recover_ring, &entry))
		schedule_work(&aer_recover_work);
	else
		pr_err("AER recover: Buffer overflow when recovering AER for %04x:%02x:%02x:%x\n",
		       domain, bus, PCI_SLOT(devfn), PCI_FUNC(devfn));
	spin_unlock_irqrestore(&aer_recover_ring_lock, flags);
}
EXPORT_SYMBOL_GPL(aer_recover_queue);

static void aer_recover_work_func(struct work_struct *work)
{
	struct aer_recover_entry entry;
	struct pci_dev *pdev;

	while (kfifo_get(&aer_recover_ring, &entry)) {
		pdev = pci_get_domain_bus_and_slot(entry.domain, entry.bus,
						   entry.devfn);
		if (!pdev) {
			pr_err("AER recover: Can not find pci_dev for %04x:%02x:%02x:%x\n",
			       entry.domain, entry.bus,
			       PCI_SLOT(entry.devfn), PCI_FUNC(entry.devfn));
			continue;
		}
		do_recovery(pdev, entry.severity);
633
		pci_dev_put(pdev);
634 635 636 637
	}
}
#endif

638 639 640 641 642 643
/**
 * get_device_error_info - read error status from dev and store it to info
 * @dev: pointer to the device expected to have a error record
 * @info: pointer to structure to store the error record
 *
 * Return 1 on success, 0 on error.
644 645
 *
 * Note that @info is reused among all error devices. Clear fields properly.
646
 */
647 648
static int get_device_error_info(struct pci_dev *dev, struct aer_err_info *info)
{
649
	int pos, temp;
650

651
	/* Must reset in this function */
652
	info->status = 0;
H
Hidetoshi Seto 已提交
653
	info->tlp_header_valid = 0;
654

655
	pos = pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR);
656 657 658

	/* The device might not support AER */
	if (!pos)
659
		return 1;
660 661 662 663

	if (info->severity == AER_CORRECTABLE) {
		pci_read_config_dword(dev, pos + PCI_ERR_COR_STATUS,
			&info->status);
664 665 666
		pci_read_config_dword(dev, pos + PCI_ERR_COR_MASK,
			&info->mask);
		if (!(info->status & ~info->mask))
667
			return 0;
668 669 670 671 672 673
	} else if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE ||
		info->severity == AER_NONFATAL) {

		/* Link is still healthy for IO reads */
		pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS,
			&info->status);
674 675 676
		pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_MASK,
			&info->mask);
		if (!(info->status & ~info->mask))
677
			return 0;
678

679 680
		/* Get First Error Pointer */
		pci_read_config_dword(dev, pos + PCI_ERR_CAP, &temp);
H
Hidetoshi Seto 已提交
681
		info->first_error = PCI_ERR_CAP_FEP(temp);
682

683
		if (info->status & AER_LOG_TLP_MASKS) {
H
Hidetoshi Seto 已提交
684
			info->tlp_header_valid = 1;
685 686 687 688 689 690 691 692 693 694 695
			pci_read_config_dword(dev,
				pos + PCI_ERR_HEADER_LOG, &info->tlp.dw0);
			pci_read_config_dword(dev,
				pos + PCI_ERR_HEADER_LOG + 4, &info->tlp.dw1);
			pci_read_config_dword(dev,
				pos + PCI_ERR_HEADER_LOG + 8, &info->tlp.dw2);
			pci_read_config_dword(dev,
				pos + PCI_ERR_HEADER_LOG + 12, &info->tlp.dw3);
		}
	}

696
	return 1;
697 698
}

Z
Zhang, Yanmin 已提交
699 700 701 702 703
static inline void aer_process_err_devices(struct pcie_device *p_device,
			struct aer_err_info *e_info)
{
	int i;

704
	/* Report all before handle them, not to lost records by reset etc. */
Z
Zhang, Yanmin 已提交
705
	for (i = 0; i < e_info->error_dev_num && e_info->dev[i]; i++) {
706
		if (get_device_error_info(e_info->dev[i], e_info))
Z
Zhang, Yanmin 已提交
707
			aer_print_error(e_info->dev[i], e_info);
708 709 710 711
	}
	for (i = 0; i < e_info->error_dev_num && e_info->dev[i]; i++) {
		if (get_device_error_info(e_info->dev[i], e_info))
			handle_error_source(p_device, e_info->dev[i], e_info);
Z
Zhang, Yanmin 已提交
712 713 714
	}
}

715 716 717 718
/**
 * aer_isr_one_error - consume an error detected by root port
 * @p_device: pointer to error root port service device
 * @e_src: pointer to an error source
R
Randy Dunlap 已提交
719
 */
720 721 722
static void aer_isr_one_error(struct pcie_device *p_device,
		struct aer_err_source *e_src)
{
723 724 725 726
	struct aer_err_info *e_info;

	/* struct aer_err_info might be big, so we allocate it with slab */
	e_info = kmalloc(sizeof(struct aer_err_info), GFP_KERNEL);
727
	if (!e_info) {
728 729 730 731
		dev_printk(KERN_DEBUG, &p_device->port->dev,
			"Can't allocate mem when processing AER errors\n");
		return;
	}
732 733 734 735 736

	/*
	 * There is a possibility that both correctable error and
	 * uncorrectable error being logged. Report correctable error first.
	 */
737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760
	if (e_src->status & PCI_ERR_ROOT_COR_RCV) {
		e_info->id = ERR_COR_ID(e_src->id);
		e_info->severity = AER_CORRECTABLE;

		if (e_src->status & PCI_ERR_ROOT_MULTI_COR_RCV)
			e_info->multi_error_valid = 1;
		else
			e_info->multi_error_valid = 0;

		aer_print_port_info(p_device->port, e_info);

		if (find_source_device(p_device->port, e_info))
			aer_process_err_devices(p_device, e_info);
	}

	if (e_src->status & PCI_ERR_ROOT_UNCOR_RCV) {
		e_info->id = ERR_UNCOR_ID(e_src->id);

		if (e_src->status & PCI_ERR_ROOT_FATAL_RCV)
			e_info->severity = AER_FATAL;
		else
			e_info->severity = AER_NONFATAL;

		if (e_src->status & PCI_ERR_ROOT_MULTI_UNCOR_RCV)
H
Hidetoshi Seto 已提交
761
			e_info->multi_error_valid = 1;
762 763
		else
			e_info->multi_error_valid = 0;
764

765 766
		aer_print_port_info(p_device->port, e_info);

767 768
		if (find_source_device(p_device->port, e_info))
			aer_process_err_devices(p_device, e_info);
769
	}
770 771

	kfree(e_info);
772 773
}

774 775 776 777 778 779 780 781 782 783 784 785 786 787 788
/**
 * get_e_source - retrieve an error source
 * @rpc: pointer to the root port which holds an error
 * @e_src: pointer to store retrieved error source
 *
 * Return 1 if an error source is retrieved, otherwise 0.
 *
 * Invoked by DPC handler to consume an error.
 */
static int get_e_source(struct aer_rpc *rpc, struct aer_err_source *e_src)
{
	unsigned long flags;

	/* Lock access to Root error producer/consumer index */
	spin_lock_irqsave(&rpc->e_lock, flags);
789 790 791
	if (rpc->prod_idx == rpc->cons_idx) {
		spin_unlock_irqrestore(&rpc->e_lock, flags);
		return 0;
792
	}
793 794 795 796 797

	*e_src = rpc->e_sources[rpc->cons_idx];
	rpc->cons_idx++;
	if (rpc->cons_idx == AER_ERROR_SOURCES_MAX)
		rpc->cons_idx = 0;
798 799
	spin_unlock_irqrestore(&rpc->e_lock, flags);

800
	return 1;
801 802
}

803 804
/**
 * aer_isr - consume errors detected by root port
805
 * @work: definition of this work item
806 807
 *
 * Invoked, as DPC, when root port records new detected error
R
Randy Dunlap 已提交
808
 */
809
void aer_isr(struct work_struct *work)
810
{
811 812
	struct aer_rpc *rpc = container_of(work, struct aer_rpc, dpc_handler);
	struct pcie_device *p_device = rpc->rpd;
813
	struct aer_err_source uninitialized_var(e_src);
814 815

	mutex_lock(&rpc->rpc_mutex);
816 817
	while (get_e_source(rpc, &e_src))
		aer_isr_one_error(p_device, &e_src);
818 819 820 821 822 823 824 825 826 827
	mutex_unlock(&rpc->rpc_mutex);

	wake_up(&rpc->wait_release);
}

/**
 * aer_init - provide AER initialization
 * @dev: pointer to AER pcie device
 *
 * Invoked when AER service driver is loaded.
R
Randy Dunlap 已提交
828
 */
829 830
int aer_init(struct pcie_device *dev)
{
831 832 833
	if (forceload) {
		dev_printk(KERN_DEBUG, &dev->device,
			   "aerdrv forceload requested.\n");
834
		pcie_aer_force_firmware_first(dev->port, 0);
835
	}
836
	return 0;
837
}