aboutsummaryrefslogtreecommitdiffstats
path: root/fs
diff options
context:
space:
mode:
authorChristoph Hellwig <hch@lst.de>2009-10-29 09:14:04 -0400
committerJens Axboe <jens.axboe@oracle.com>2009-10-29 09:14:04 -0400
commitab0a9735e06914ce4d2a94ffa41497dbc142fe7f (patch)
tree4d3a19c066d2e513f7d1645092b4a27059b5ee67 /fs
parentb9d128f1088ea5245109dfc9bbceb128b6371a77 (diff)
blkdev: flush disk cache on ->fsync
Currently there is no barrier support in the block device code. That means we cannot guarantee any sort of data integerity when using the block device node with dis kwrite caches enabled. Using the raw block device node is a typical use case for virtualization (and I assume databases, too). This patch changes block_fsync to issue a cache flush and thus make fsync on block device nodes actually useful. Note that in mainline we would also need to add such code to the ->aio_write method for O_SYNC handling, but assuming that Jan's patch series for the O_SYNC rewrite goes in it will also call into ->fsync for 2.6.32. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Diffstat (limited to 'fs')
-rw-r--r--fs/block_dev.c12
1 files changed, 11 insertions, 1 deletions
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 9cf4b926f8e4..dde91e7e1c3a 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -405,7 +405,17 @@ static loff_t block_llseek(struct file *file, loff_t offset, int origin)
405 405
406static int block_fsync(struct file *filp, struct dentry *dentry, int datasync) 406static int block_fsync(struct file *filp, struct dentry *dentry, int datasync)
407{ 407{
408 return sync_blockdev(I_BDEV(filp->f_mapping->host)); 408 struct block_device *bdev = I_BDEV(filp->f_mapping->host);
409 int error;
410
411 error = sync_blockdev(bdev);
412 if (error)
413 return error;
414
415 error = blkdev_issue_flush(bdev, NULL);
416 if (error == -EOPNOTSUPP)
417 error = 0;
418 return error;
409} 419}
410 420
411/* 421/*
48'>448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640
/*
 *  acpi_power.c - ACPI Bus Power Management ($Revision: 39 $)
 *
 *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com>
 *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
 *
 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 *
 *  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, write to the Free Software Foundation, Inc.,
 *  59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.
 *
 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 */

/*
 * ACPI power-managed devices may be controlled in two ways:
 * 1. via "Device Specific (D-State) Control"
 * 2. via "Power Resource Control".
 * This module is used to manage devices relying on Power Resource Control.
 * 
 * An ACPI "power resource object" describes a software controllable power
 * plane, clock plane, or other resource used by a power managed device.
 * A device may rely on multiple power resources, and a power resource
 * may be shared by multiple devices.
 */

#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/types.h>
#include <linux/slab.h>
#include <acpi/acpi_bus.h>
#include <acpi/acpi_drivers.h>
#include "sleep.h"

#define PREFIX "ACPI: "

#define _COMPONENT			ACPI_POWER_COMPONENT
ACPI_MODULE_NAME("power");
#define ACPI_POWER_CLASS		"power_resource"
#define ACPI_POWER_DEVICE_NAME		"Power Resource"
#define ACPI_POWER_FILE_INFO		"info"
#define ACPI_POWER_FILE_STATUS		"state"
#define ACPI_POWER_RESOURCE_STATE_OFF	0x00
#define ACPI_POWER_RESOURCE_STATE_ON	0x01
#define ACPI_POWER_RESOURCE_STATE_UNKNOWN 0xFF

static int acpi_power_add(struct acpi_device *device);
static int acpi_power_remove(struct acpi_device *device, int type);
static int acpi_power_resume(struct acpi_device *device);

static const struct acpi_device_id power_device_ids[] = {
	{ACPI_POWER_HID, 0},
	{"", 0},
};
MODULE_DEVICE_TABLE(acpi, power_device_ids);

static struct acpi_driver acpi_power_driver = {
	.name = "power",
	.class = ACPI_POWER_CLASS,
	.ids = power_device_ids,
	.ops = {
		.add = acpi_power_add,
		.remove = acpi_power_remove,
		.resume = acpi_power_resume,
		},
};

struct acpi_power_resource {
	struct acpi_device * device;
	acpi_bus_id name;
	u32 system_level;
	u32 order;
	unsigned int ref_count;
	struct mutex resource_lock;
};

static struct list_head acpi_power_resource_list;

/* --------------------------------------------------------------------------
                             Power Resource Management
   -------------------------------------------------------------------------- */

static int
acpi_power_get_context(acpi_handle handle,
		       struct acpi_power_resource **resource)
{
	int result = 0;
	struct acpi_device *device = NULL;


	if (!resource)
		return -ENODEV;

	result = acpi_bus_get_device(handle, &device);
	if (result) {
		printk(KERN_WARNING PREFIX "Getting context [%p]\n", handle);
		return result;
	}

	*resource = acpi_driver_data(device);
	if (!*resource)
		return -ENODEV;

	return 0;
}

static int acpi_power_get_state(acpi_handle handle, int *state)
{
	acpi_status status = AE_OK;
	unsigned long long sta = 0;
	char node_name[5];
	struct acpi_buffer buffer = { sizeof(node_name), node_name };


	if (!handle || !state)
		return -EINVAL;

	status = acpi_evaluate_integer(handle, "_STA", NULL, &sta);
	if (ACPI_FAILURE(status))
		return -ENODEV;

	*state = (sta & 0x01)?ACPI_POWER_RESOURCE_STATE_ON:
			      ACPI_POWER_RESOURCE_STATE_OFF;

	acpi_get_name(handle, ACPI_SINGLE_NAME, &buffer);

	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Resource [%s] is %s\n",
			  node_name,
				*state ? "on" : "off"));

	return 0;
}

static int acpi_power_get_list_state(struct acpi_handle_list *list, int *state)
{
	int cur_state;
	int i = 0;

	if (!list || !state)
		return -EINVAL;

	/* The state of the list is 'on' IFF all resources are 'on'. */

	for (i = 0; i < list->count; i++) {
		struct acpi_power_resource *resource;
		acpi_handle handle = list->handles[i];
		int result;

		result = acpi_power_get_context(handle, &resource);
		if (result)
			return result;

		mutex_lock(&resource->resource_lock);

		result = acpi_power_get_state(handle, &cur_state);

		mutex_unlock(&resource->resource_lock);

		if (result)
			return result;

		if (cur_state != ACPI_POWER_RESOURCE_STATE_ON)
			break;
	}

	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Resource list is %s\n",
			  cur_state ? "on" : "off"));

	*state = cur_state;

	return 0;
}

static int __acpi_power_on(struct acpi_power_resource *resource)
{
	acpi_status status = AE_OK;

	status = acpi_evaluate_object(resource->device->handle, "_ON", NULL, NULL);
	if (ACPI_FAILURE(status))
		return -ENODEV;

	/* Update the power resource's _device_ power state */
	resource->device->power.state = ACPI_STATE_D0;

	ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Power resource [%s] turned on\n",
			  resource->name));

	return 0;
}

static int acpi_power_on(acpi_handle handle)
{
	int result = 0;
	struct acpi_power_resource *resource = NULL;

	result = acpi_power_get_context(handle, &resource);
	if (result)
		return result;

	mutex_lock(&resource->resource_lock);

	if (resource->ref_count++) {
		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
				  "Power resource [%s] already on",
				  resource->name));
	} else {
		result = __acpi_power_on(resource);
		if (result)
			resource->ref_count--;
	}

	mutex_unlock(&resource->resource_lock);

	return result;
}

static int acpi_power_off(acpi_handle handle)
{
	int result = 0;
	acpi_status status = AE_OK;
	struct acpi_power_resource *resource = NULL;

	result = acpi_power_get_context(handle, &resource);
	if (result)
		return result;

	mutex_lock(&resource->resource_lock);

	if (!resource->ref_count) {
		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
				  "Power resource [%s] already off",
				  resource->name));
		goto unlock;
	}

	if (--resource->ref_count) {
		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
				  "Power resource [%s] still in use\n",
				  resource->name));
		goto unlock;
	}

	status = acpi_evaluate_object(resource->device->handle, "_OFF", NULL, NULL);
	if (ACPI_FAILURE(status)) {
		result = -ENODEV;
	} else {
		/* Update the power resource's _device_ power state */
		resource->device->power.state = ACPI_STATE_D3;

		ACPI_DEBUG_PRINT((ACPI_DB_INFO,
				  "Power resource [%s] turned off\n",
				  resource->name));
	}

 unlock:
	mutex_unlock(&resource->resource_lock);

	return result;
}

static void __acpi_power_off_list(struct acpi_handle_list *list, int num_res)
{
	int i;

	for (i = num_res - 1; i >= 0 ; i--)
		acpi_power_off(list->handles[i]);
}

static void acpi_power_off_list(struct acpi_handle_list *list)
{
	__acpi_power_off_list(list, list->count);
}

static int acpi_power_on_list(struct acpi_handle_list *list)
{
	int result = 0;
	int i;

	for (i = 0; i < list->count; i++) {
		result = acpi_power_on(list->handles[i]);
		if (result) {
			__acpi_power_off_list(list, i);
			break;
		}
	}

	return result;
}

/**
 * acpi_device_sleep_wake - execute _DSW (Device Sleep Wake) or (deprecated in
 *                          ACPI 3.0) _PSW (Power State Wake)
 * @dev: Device to handle.
 * @enable: 0 - disable, 1 - enable the wake capabilities of the device.
 * @sleep_state: Target sleep state of the system.
 * @dev_state: Target power state of the device.
 *
 * Execute _DSW (Device Sleep Wake) or (deprecated in ACPI 3.0) _PSW (Power
 * State Wake) for the device, if present.  On failure reset the device's
 * wakeup.flags.valid flag.
 *
 * RETURN VALUE:
 * 0 if either _DSW or _PSW has been successfully executed
 * 0 if neither _DSW nor _PSW has been found
 * -ENODEV if the execution of either _DSW or _PSW has failed
 */
int acpi_device_sleep_wake(struct acpi_device *dev,
                           int enable, int sleep_state, int dev_state)
{
	union acpi_object in_arg[3];
	struct acpi_object_list arg_list = { 3, in_arg };
	acpi_status status = AE_OK;

	/*
	 * Try to execute _DSW first.
	 *
	 * Three agruments are needed for the _DSW object:
	 * Argument 0: enable/disable the wake capabilities
	 * Argument 1: target system state
	 * Argument 2: target device state
	 * When _DSW object is called to disable the wake capabilities, maybe
	 * the first argument is filled. The values of the other two agruments
	 * are meaningless.
	 */
	in_arg[0].type = ACPI_TYPE_INTEGER;
	in_arg[0].integer.value = enable;
	in_arg[1].type = ACPI_TYPE_INTEGER;
	in_arg[1].integer.value = sleep_state;
	in_arg[2].type = ACPI_TYPE_INTEGER;
	in_arg[2].integer.value = dev_state;
	status = acpi_evaluate_object(dev->handle, "_DSW", &arg_list, NULL);
	if (ACPI_SUCCESS(status)) {
		return 0;
	} else if (status != AE_NOT_FOUND) {
		printk(KERN_ERR PREFIX "_DSW execution failed\n");
		dev->wakeup.flags.valid = 0;
		return -ENODEV;
	}

	/* Execute _PSW */
	arg_list.count = 1;
	in_arg[0].integer.value = enable;
	status = acpi_evaluate_object(dev->handle, "_PSW", &arg_list, NULL);
	if (ACPI_FAILURE(status) && (status != AE_NOT_FOUND)) {
		printk(KERN_ERR PREFIX "_PSW execution failed\n");
		dev->wakeup.flags.valid = 0;
		return -ENODEV;
	}

	return 0;
}

/*
 * Prepare a wakeup device, two steps (Ref ACPI 2.0:P229):
 * 1. Power on the power resources required for the wakeup device 
 * 2. Execute _DSW (Device Sleep Wake) or (deprecated in ACPI 3.0) _PSW (Power
 *    State Wake) for the device, if present
 */
int acpi_enable_wakeup_device_power(struct acpi_device *dev, int sleep_state)
{
	int i, err = 0;

	if (!dev || !dev->wakeup.flags.valid)
		return -EINVAL;

	mutex_lock(&acpi_device_lock);

	if (dev->wakeup.prepare_count++)
		goto out;

	/* Open power resource */
	for (i = 0; i < dev->wakeup.resources.count; i++) {
		int ret = acpi_power_on(dev->wakeup.resources.handles[i]);
		if (ret) {
			printk(KERN_ERR PREFIX "Transition power state\n");
			dev->wakeup.flags.valid = 0;
			err = -ENODEV;
			goto err_out;
		}
	}

	/*
	 * Passing 3 as the third argument below means the device may be placed
	 * in arbitrary power state afterwards.
	 */
	err = acpi_device_sleep_wake(dev, 1, sleep_state, 3);

 err_out:
	if (err)
		dev->wakeup.prepare_count = 0;

 out:
	mutex_unlock(&acpi_device_lock);
	return err;
}

/*
 * Shutdown a wakeup device, counterpart of above method
 * 1. Execute _DSW (Device Sleep Wake) or (deprecated in ACPI 3.0) _PSW (Power
 *    State Wake) for the device, if present
 * 2. Shutdown down the power resources
 */
int acpi_disable_wakeup_device_power(struct acpi_device *dev)
{
	int i, err = 0;

	if (!dev || !dev->wakeup.flags.valid)
		return -EINVAL;

	mutex_lock(&acpi_device_lock);

	if (--dev->wakeup.prepare_count > 0)
		goto out;

	/*
	 * Executing the code below even if prepare_count is already zero when
	 * the function is called may be useful, for example for initialisation.
	 */
	if (dev->wakeup.prepare_count < 0)
		dev->wakeup.prepare_count = 0;

	err = acpi_device_sleep_wake(dev, 0, 0, 0);
	if (err)
		goto out;

	/* Close power resource */
	for (i = 0; i < dev->wakeup.resources.count; i++) {
		int ret = acpi_power_off(dev->wakeup.resources.handles[i]);
		if (ret) {
			printk(KERN_ERR PREFIX "Transition power state\n");
			dev->wakeup.flags.valid = 0;
			err = -ENODEV;
			goto out;
		}
	}

 out:
	mutex_unlock(&acpi_device_lock);
	return err;
}

/* --------------------------------------------------------------------------
                             Device Power Management
   -------------------------------------------------------------------------- */

int acpi_power_get_inferred_state(struct acpi_device *device, int *state)
{
	int result = 0;
	struct acpi_handle_list *list = NULL;
	int list_state = 0;
	int i = 0;

	if (!device || !state)
		return -EINVAL;

	/*
	 * We know a device's inferred power state when all the resources
	 * required for a given D-state are 'on'.
	 */
	for (i = ACPI_STATE_D0; i < ACPI_STATE_D3; i++) {
		list = &device->power.states[i].resources;
		if (list->count < 1)
			continue;

		result = acpi_power_get_list_state(list, &list_state);
		if (result)
			return result;

		if (list_state == ACPI_POWER_RESOURCE_STATE_ON) {
			*state = i;
			return 0;
		}
	}

	*state = ACPI_STATE_D3;
	return 0;
}

int acpi_power_on_resources(struct acpi_device *device, int state)
{
	if (!device || state < ACPI_STATE_D0 || state > ACPI_STATE_D3)
		return -EINVAL;

	return acpi_power_on_list(&device->power.states[state].resources);
}

int acpi_power_transition(struct acpi_device *device, int state)
{
	int result;

	if (!device || (state < ACPI_STATE_D0) || (state > ACPI_STATE_D3))
		return -EINVAL;

	if (device->power.state == state)
		return 0;

	if ((device->power.state < ACPI_STATE_D0)
	    || (device->power.state > ACPI_STATE_D3))
		return -ENODEV;

	/* TBD: Resources must be ordered. */

	/*
	 * First we reference all power resources required in the target list
	 * (e.g. so the device doesn't lose power while transitioning).  Then,
	 * we dereference all power resources used in the current list.
	 */
	result = acpi_power_on_list(&device->power.states[state].resources);
	if (!result)
		acpi_power_off_list(
			&device->power.states[device->power.state].resources);

	/* We shouldn't change the state unless the above operations succeed. */
	device->power.state = result ? ACPI_STATE_UNKNOWN : state;

	return result;
}