Eric Lee / linux-smarc-t335x-v3.2

Commit b8887e6e8c04bcefb512cdb08fc7e9c310ac847e

Authored by Randy Dunlap 2005-11-07 17:01:07 +0800

Committed by Linus Torvalds 2005-11-07 23:53:55 +0800

Exists in master and in 4 other branches

[PATCH] kernel-docs: fix kernel-doc format problems

Convert to proper kernel-doc format.

Some have extra blank lines (not allowed immed.  after the function name)
or need blank lines (after all parameters).  Function summary must be only
one line.

Colon (":") in a function description does weird things (causes kernel-doc
to think that it's a new section head sadly).

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Showing 4 changed files with 4 additions and 5 deletions Inline Diff

drivers/block/ll_rw_blk.c
fs/fs-writeback.c
include/linux/kernel.h
kernel/sys.c

drivers/block/ll_rw_blk.c

Diff comments View file @ b8887e6

 /*
  *  linux/drivers/block/ll_rw_blk.c
  *
  * Copyright (C) 1991, 1992 Linus Torvalds
  * Copyright (C) 1994,      Karl Keyte: Added support for disk statistics
  * Elevator latency, (C) 2000  Andrea Arcangeli <andrea@suse.de> SuSE
  * Queue request tables / lock, selectable elevator, Jens Axboe <axboe@suse.de>
  * kernel-doc documentation started by NeilBrown <neilb@cse.unsw.edu.au> -  July2000
  * bio rewrite, highmem i/o, etc, Jens Axboe <axboe@suse.de> - may 2001
  */
 /*
  * This handles all read/write requests to block devices
  */
 #include <linux/config.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/backing-dev.h>
 #include <linux/bio.h>
 #include <linux/blkdev.h>
 #include <linux/highmem.h>
 #include <linux/mm.h>
 #include <linux/kernel_stat.h>
 #include <linux/string.h>
 #include <linux/init.h>
 #include <linux/bootmem.h>	/* for max_pfn/max_low_pfn */
 #include <linux/completion.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
 /*
  * for max sense size
  */
 #include <scsi/scsi_cmnd.h>
 static void blk_unplug_work(void *data);
 static void blk_unplug_timeout(unsigned long data);
 static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io);
 /*
  * For the allocated request tables
  */
 static kmem_cache_t *request_cachep;
 /*
  * For queue allocation
  */
 static kmem_cache_t *requestq_cachep;
 /*
  * For io context allocations
  */
 static kmem_cache_t *iocontext_cachep;
 static wait_queue_head_t congestion_wqh[2] = {
 		__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[0]),
 		__WAIT_QUEUE_HEAD_INITIALIZER(congestion_wqh[1])
 	};
 /*
  * Controlling structure to kblockd
  */
 static struct workqueue_struct *kblockd_workqueue;
 unsigned long blk_max_low_pfn, blk_max_pfn;
 EXPORT_SYMBOL(blk_max_low_pfn);
 EXPORT_SYMBOL(blk_max_pfn);
 /* Amount of time in which a process may batch requests */
 #define BLK_BATCH_TIME	(HZ/50UL)
 /* Number of requests a "batching" process may submit */
 #define BLK_BATCH_REQ	32
 /*
  * Return the threshold (number of used requests) at which the queue is
  * considered to be congested.  It include a little hysteresis to keep the
  * context switch rate down.
  */
 static inline int queue_congestion_on_threshold(struct request_queue *q)
 {
 	return q->nr_congestion_on;
 }
 /*
  * The threshold at which a queue is considered to be uncongested
  */
 static inline int queue_congestion_off_threshold(struct request_queue *q)
 {
 	return q->nr_congestion_off;
 }
 static void blk_queue_congestion_threshold(struct request_queue *q)
 {
 	int nr;
 	nr = q->nr_requests - (q->nr_requests / 8) + 1;
 	if (nr > q->nr_requests)
 		nr = q->nr_requests;
 	q->nr_congestion_on = nr;
 	nr = q->nr_requests - (q->nr_requests / 8) - (q->nr_requests / 16) - 1;
 	if (nr < 1)
 		nr = 1;
 	q->nr_congestion_off = nr;
 }
 /*
  * A queue has just exitted congestion.  Note this in the global counter of
  * congested queues, and wake up anyone who was waiting for requests to be
  * put back.
  */
 static void clear_queue_congested(request_queue_t *q, int rw)
 {
 	enum bdi_state bit;
 	wait_queue_head_t *wqh = &congestion_wqh[rw];
 	bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested;
 	clear_bit(bit, &q->backing_dev_info.state);
 	smp_mb__after_clear_bit();
 	if (waitqueue_active(wqh))
 		wake_up(wqh);
 }
 /*
  * A queue has just entered congestion.  Flag that in the queue's VM-visible
  * state flags and increment the global gounter of congested queues.
  */
 static void set_queue_congested(request_queue_t *q, int rw)
 {
 	enum bdi_state bit;
 	bit = (rw == WRITE) ? BDI_write_congested : BDI_read_congested;
 	set_bit(bit, &q->backing_dev_info.state);
 }
 /**
  * blk_get_backing_dev_info - get the address of a queue's backing_dev_info
  * @bdev:	device
  *
  * Locates the passed device's request queue and returns the address of its
  * backing_dev_info
  *
  * Will return NULL if the request queue cannot be located.
  */
 struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev)
 {
 	struct backing_dev_info *ret = NULL;
 	request_queue_t *q = bdev_get_queue(bdev);
 	if (q)
 		ret = &q->backing_dev_info;
 	return ret;
 }
 EXPORT_SYMBOL(blk_get_backing_dev_info);
 void blk_queue_activity_fn(request_queue_t *q, activity_fn *fn, void *data)
 {
 	q->activity_fn = fn;
 	q->activity_data = data;
 }
 EXPORT_SYMBOL(blk_queue_activity_fn);
 /**
  * blk_queue_prep_rq - set a prepare_request function for queue
  * @q:		queue
  * @pfn:	prepare_request function
  *
  * It's possible for a queue to register a prepare_request callback which
  * is invoked before the request is handed to the request_fn. The goal of
  * the function is to prepare a request for I/O, it can be used to build a
  * cdb from the request data for instance.
  *
  */
 void blk_queue_prep_rq(request_queue_t *q, prep_rq_fn *pfn)
 {
 	q->prep_rq_fn = pfn;
 }
 EXPORT_SYMBOL(blk_queue_prep_rq);
 /**
  * blk_queue_merge_bvec - set a merge_bvec function for queue
  * @q:		queue
  * @mbfn:	merge_bvec_fn
  *
  * Usually queues have static limitations on the max sectors or segments that
  * we can put in a request. Stacking drivers may have some settings that
  * are dynamic, and thus we have to query the queue whether it is ok to
  * add a new bio_vec to a bio at a given offset or not. If the block device
  * has such limitations, it needs to register a merge_bvec_fn to control
  * the size of bio's sent to it. Note that a block device *must* allow a
  * single page to be added to an empty bio. The block device driver may want
  * to use the bio_split() function to deal with these bio's. By default
  * no merge_bvec_fn is defined for a queue, and only the fixed limits are
  * honored.
  */
 void blk_queue_merge_bvec(request_queue_t *q, merge_bvec_fn *mbfn)
 {
 	q->merge_bvec_fn = mbfn;
 }
 EXPORT_SYMBOL(blk_queue_merge_bvec);
 /**
  * blk_queue_make_request - define an alternate make_request function for a device
  * @q:  the request queue for the device to be affected
  * @mfn: the alternate make_request function
  *
  * Description:
  *    The normal way for &struct bios to be passed to a device
  *    driver is for them to be collected into requests on a request
  *    queue, and then to allow the device driver to select requests
  *    off that queue when it is ready.  This works well for many block
  *    devices. However some block devices (typically virtual devices
  *    such as md or lvm) do not benefit from the processing on the
  *    request queue, and are served best by having the requests passed
  *    directly to them.  This can be achieved by providing a function
  *    to blk_queue_make_request().
  *
  * Caveat:
  *    The driver that does this *must* be able to deal appropriately
  *    with buffers in "highmemory". This can be accomplished by either calling
  *    __bio_kmap_atomic() to get a temporary kernel mapping, or by calling
  *    blk_queue_bounce() to create a buffer in normal memory.
  **/
 void blk_queue_make_request(request_queue_t * q, make_request_fn * mfn)
 {
 	/*
 	 * set defaults
 	 */
 	q->nr_requests = BLKDEV_MAX_RQ;
 	blk_queue_max_phys_segments(q, MAX_PHYS_SEGMENTS);
 	blk_queue_max_hw_segments(q, MAX_HW_SEGMENTS);
 	q->make_request_fn = mfn;
 	q->backing_dev_info.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE;
 	q->backing_dev_info.state = 0;
 	q->backing_dev_info.capabilities = BDI_CAP_MAP_COPY;
 	blk_queue_max_sectors(q, MAX_SECTORS);
 	blk_queue_hardsect_size(q, 512);
 	blk_queue_dma_alignment(q, 511);
 	blk_queue_congestion_threshold(q);
 	q->nr_batching = BLK_BATCH_REQ;
 	q->unplug_thresh = 4;		/* hmm */
 	q->unplug_delay = (3 * HZ) / 1000;	/* 3 milliseconds */
 	if (q->unplug_delay == 0)
 		q->unplug_delay = 1;
 	INIT_WORK(&q->unplug_work, blk_unplug_work, q);
 	q->unplug_timer.function = blk_unplug_timeout;
 	q->unplug_timer.data = (unsigned long)q;
 	/*
 	 * by default assume old behaviour and bounce for any highmem page
 	 */
 	blk_queue_bounce_limit(q, BLK_BOUNCE_HIGH);
 	blk_queue_activity_fn(q, NULL, NULL);
 }
 EXPORT_SYMBOL(blk_queue_make_request);
 static inline void rq_init(request_queue_t *q, struct request *rq)
 {
 	INIT_LIST_HEAD(&rq->queuelist);
 	rq->errors = 0;
 	rq->rq_status = RQ_ACTIVE;
 	rq->bio = rq->biotail = NULL;
 	rq->ioprio = 0;
 	rq->buffer = NULL;
 	rq->ref_count = 1;
 	rq->q = q;
 	rq->waiting = NULL;
 	rq->special = NULL;
 	rq->data_len = 0;
 	rq->data = NULL;
 	rq->nr_phys_segments = 0;
 	rq->sense = NULL;
 	rq->end_io = NULL;
 	rq->end_io_data = NULL;
 }
 /**
  * blk_queue_ordered - does this queue support ordered writes
  * @q:     the request queue
  * @flag:  see below
  *
  * Description:
  *   For journalled file systems, doing ordered writes on a commit
  *   block instead of explicitly doing wait_on_buffer (which is bad
  *   for performance) can be a big win. Block drivers supporting this
  *   feature should call this function and indicate so.
  *
  **/
 void blk_queue_ordered(request_queue_t *q, int flag)
 {
 	switch (flag) {
 		case QUEUE_ORDERED_NONE:
 			if (q->flush_rq)
 				kmem_cache_free(request_cachep, q->flush_rq);
 			q->flush_rq = NULL;
 			q->ordered = flag;
 			break;
 		case QUEUE_ORDERED_TAG:
 			q->ordered = flag;
 			break;
 		case QUEUE_ORDERED_FLUSH:
 			q->ordered = flag;
 			if (!q->flush_rq)
 				q->flush_rq = kmem_cache_alloc(request_cachep,
 								GFP_KERNEL);
 			break;
 		default:
 			printk("blk_queue_ordered: bad value %d\n", flag);
 			break;
 	}
 }
 EXPORT_SYMBOL(blk_queue_ordered);
 /**
  * blk_queue_issue_flush_fn - set function for issuing a flush
  * @q:     the request queue
  * @iff:   the function to be called issuing the flush
  *
  * Description:
  *   If a driver supports issuing a flush command, the support is notified
  *   to the block layer by defining it through this call.
  *
  **/
 void blk_queue_issue_flush_fn(request_queue_t *q, issue_flush_fn *iff)
 {
 	q->issue_flush_fn = iff;
 }
 EXPORT_SYMBOL(blk_queue_issue_flush_fn);
 /*
  * Cache flushing for ordered writes handling
  */
 static void blk_pre_flush_end_io(struct request *flush_rq)
 {
 	struct request *rq = flush_rq->end_io_data;
 	request_queue_t *q = rq->q;
 	elv_completed_request(q, flush_rq);
 	rq->flags |= REQ_BAR_PREFLUSH;
 	if (!flush_rq->errors)
 		elv_requeue_request(q, rq);
 	else {
 		q->end_flush_fn(q, flush_rq);
 		clear_bit(QUEUE_FLAG_FLUSH, &q->queue_flags);
 		q->request_fn(q);
 	}
 }
 static void blk_post_flush_end_io(struct request *flush_rq)
 {
 	struct request *rq = flush_rq->end_io_data;
 	request_queue_t *q = rq->q;
 	elv_completed_request(q, flush_rq);
 	rq->flags |= REQ_BAR_POSTFLUSH;
 	q->end_flush_fn(q, flush_rq);
 	clear_bit(QUEUE_FLAG_FLUSH, &q->queue_flags);
 	q->request_fn(q);
 }
 struct request *blk_start_pre_flush(request_queue_t *q, struct request *rq)
 {
 	struct request *flush_rq = q->flush_rq;
 	BUG_ON(!blk_barrier_rq(rq));
 	if (test_and_set_bit(QUEUE_FLAG_FLUSH, &q->queue_flags))
 		return NULL;
 	rq_init(q, flush_rq);
 	flush_rq->elevator_private = NULL;
 	flush_rq->flags = REQ_BAR_FLUSH;
 	flush_rq->rq_disk = rq->rq_disk;
 	flush_rq->rl = NULL;
 	/*
 	 * prepare_flush returns 0 if no flush is needed, just mark both
 	 * pre and post flush as done in that case
 	 */
 	if (!q->prepare_flush_fn(q, flush_rq)) {
 		rq->flags |= REQ_BAR_PREFLUSH | REQ_BAR_POSTFLUSH;
 		clear_bit(QUEUE_FLAG_FLUSH, &q->queue_flags);
 		return rq;
 	}
 	/*
 	 * some drivers dequeue requests right away, some only after io
 	 * completion. make sure the request is dequeued.
 	 */
 	if (!list_empty(&rq->queuelist))
 		blkdev_dequeue_request(rq);
 	flush_rq->end_io_data = rq;
 	flush_rq->end_io = blk_pre_flush_end_io;
 	__elv_add_request(q, flush_rq, ELEVATOR_INSERT_FRONT, 0);
 	return flush_rq;
 }
 static void blk_start_post_flush(request_queue_t *q, struct request *rq)
 {
 	struct request *flush_rq = q->flush_rq;
 	BUG_ON(!blk_barrier_rq(rq));
 	rq_init(q, flush_rq);
 	flush_rq->elevator_private = NULL;
 	flush_rq->flags = REQ_BAR_FLUSH;
 	flush_rq->rq_disk = rq->rq_disk;
 	flush_rq->rl = NULL;
 	if (q->prepare_flush_fn(q, flush_rq)) {
 		flush_rq->end_io_data = rq;
 		flush_rq->end_io = blk_post_flush_end_io;
 		__elv_add_request(q, flush_rq, ELEVATOR_INSERT_FRONT, 0);
 		q->request_fn(q);
 	}
 }
 static inline int blk_check_end_barrier(request_queue_t *q, struct request *rq,
 					int sectors)
 {
 	if (sectors > rq->nr_sectors)
 		sectors = rq->nr_sectors;
 	rq->nr_sectors -= sectors;
 	return rq->nr_sectors;
 }
 static int __blk_complete_barrier_rq(request_queue_t *q, struct request *rq,
 				     int sectors, int queue_locked)
 {
 	if (q->ordered != QUEUE_ORDERED_FLUSH)
 		return 0;
 	if (!blk_fs_request(rq) || !blk_barrier_rq(rq))
 		return 0;
 	if (blk_barrier_postflush(rq))
 		return 0;
 	if (!blk_check_end_barrier(q, rq, sectors)) {
 		unsigned long flags = 0;
 		if (!queue_locked)
 			spin_lock_irqsave(q->queue_lock, flags);
 		blk_start_post_flush(q, rq);
 		if (!queue_locked)
 			spin_unlock_irqrestore(q->queue_lock, flags);
 	}
 	return 1;
 }
 /**
  * blk_complete_barrier_rq - complete possible barrier request
  * @q:  the request queue for the device
  * @rq:  the request
  * @sectors:  number of sectors to complete
  *
  * Description:
  *   Used in driver end_io handling to determine whether to postpone
  *   completion of a barrier request until a post flush has been done. This
  *   is the unlocked variant, used if the caller doesn't already hold the
  *   queue lock.
  **/
 int blk_complete_barrier_rq(request_queue_t *q, struct request *rq, int sectors)
 {
 	return __blk_complete_barrier_rq(q, rq, sectors, 0);
 }
 EXPORT_SYMBOL(blk_complete_barrier_rq);
 /**
  * blk_complete_barrier_rq_locked - complete possible barrier request
  * @q:  the request queue for the device
  * @rq:  the request
  * @sectors:  number of sectors to complete
  *
  * Description:
  *   See blk_complete_barrier_rq(). This variant must be used if the caller
  *   holds the queue lock.
  **/
 int blk_complete_barrier_rq_locked(request_queue_t *q, struct request *rq,
 				   int sectors)
 {
 	return __blk_complete_barrier_rq(q, rq, sectors, 1);
 }
 EXPORT_SYMBOL(blk_complete_barrier_rq_locked);
 /**
  * blk_queue_bounce_limit - set bounce buffer limit for queue
  * @q:  the request queue for the device
  * @dma_addr:   bus address limit
  *
  * Description:
  *    Different hardware can have different requirements as to what pages
  *    it can do I/O directly to. A low level driver can call
  *    blk_queue_bounce_limit to have lower memory pages allocated as bounce
  *    buffers for doing I/O to pages residing above @page. By default
  *    the block layer sets this to the highest numbered "low" memory page.
  **/
 void blk_queue_bounce_limit(request_queue_t *q, u64 dma_addr)
 {
 	unsigned long bounce_pfn = dma_addr >> PAGE_SHIFT;
 	/*
 	 * set appropriate bounce gfp mask -- unfortunately we don't have a
 	 * full 4GB zone, so we have to resort to low memory for any bounces.
 	 * ISA has its own < 16MB zone.
 	 */
 	if (bounce_pfn < blk_max_low_pfn) {
 		BUG_ON(dma_addr < BLK_BOUNCE_ISA);
 		init_emergency_isa_pool();
 		q->bounce_gfp = GFP_NOIO | GFP_DMA;
 	} else
 		q->bounce_gfp = GFP_NOIO;
 	q->bounce_pfn = bounce_pfn;
 }
 EXPORT_SYMBOL(blk_queue_bounce_limit);
 /**
  * blk_queue_max_sectors - set max sectors for a request for this queue
  * @q:  the request queue for the device
  * @max_sectors:  max sectors in the usual 512b unit
  *
  * Description:
  *    Enables a low level driver to set an upper limit on the size of
  *    received requests.
  **/
 void blk_queue_max_sectors(request_queue_t *q, unsigned short max_sectors)
 {
 	if ((max_sectors << 9) < PAGE_CACHE_SIZE) {
 		max_sectors = 1 << (PAGE_CACHE_SHIFT - 9);
 		printk("%s: set to minimum %d\n", __FUNCTION__, max_sectors);
 	}
 	q->max_sectors = q->max_hw_sectors = max_sectors;
 }
 EXPORT_SYMBOL(blk_queue_max_sectors);
 /**
  * blk_queue_max_phys_segments - set max phys segments for a request for this queue
  * @q:  the request queue for the device
  * @max_segments:  max number of segments
  *
  * Description:
  *    Enables a low level driver to set an upper limit on the number of
  *    physical data segments in a request.  This would be the largest sized
  *    scatter list the driver could handle.
  **/
 void blk_queue_max_phys_segments(request_queue_t *q, unsigned short max_segments)
 {
 	if (!max_segments) {
 		max_segments = 1;
 		printk("%s: set to minimum %d\n", __FUNCTION__, max_segments);
 	}
 	q->max_phys_segments = max_segments;
 }
 EXPORT_SYMBOL(blk_queue_max_phys_segments);
 /**
  * blk_queue_max_hw_segments - set max hw segments for a request for this queue
  * @q:  the request queue for the device
  * @max_segments:  max number of segments
  *
  * Description:
  *    Enables a low level driver to set an upper limit on the number of
  *    hw data segments in a request.  This would be the largest number of
  *    address/length pairs the host adapter can actually give as once
  *    to the device.
  **/
 void blk_queue_max_hw_segments(request_queue_t *q, unsigned short max_segments)
 {
 	if (!max_segments) {
 		max_segments = 1;
 		printk("%s: set to minimum %d\n", __FUNCTION__, max_segments);
 	}
 	q->max_hw_segments = max_segments;
 }
 EXPORT_SYMBOL(blk_queue_max_hw_segments);
 /**
  * blk_queue_max_segment_size - set max segment size for blk_rq_map_sg
  * @q:  the request queue for the device
  * @max_size:  max size of segment in bytes
  *
  * Description:
  *    Enables a low level driver to set an upper limit on the size of a
  *    coalesced segment
  **/
 void blk_queue_max_segment_size(request_queue_t *q, unsigned int max_size)
 {
 	if (max_size < PAGE_CACHE_SIZE) {
 		max_size = PAGE_CACHE_SIZE;
 		printk("%s: set to minimum %d\n", __FUNCTION__, max_size);
 	}
 	q->max_segment_size = max_size;
 }
 EXPORT_SYMBOL(blk_queue_max_segment_size);
 /**
  * blk_queue_hardsect_size - set hardware sector size for the queue
  * @q:  the request queue for the device
  * @size:  the hardware sector size, in bytes
  *
  * Description:
  *   This should typically be set to the lowest possible sector size
  *   that the hardware can operate on (possible without reverting to
  *   even internal read-modify-write operations). Usually the default
  *   of 512 covers most hardware.
  **/
 void blk_queue_hardsect_size(request_queue_t *q, unsigned short size)
 {
 	q->hardsect_size = size;
 }
 EXPORT_SYMBOL(blk_queue_hardsect_size);
 /*
  * Returns the minimum that is _not_ zero, unless both are zero.
  */
 #define min_not_zero(l, r) (l == 0) ? r : ((r == 0) ? l : min(l, r))
 /**
  * blk_queue_stack_limits - inherit underlying queue limits for stacked drivers
  * @t:	the stacking driver (top)
  * @b:  the underlying device (bottom)
  **/
 void blk_queue_stack_limits(request_queue_t *t, request_queue_t *b)
 {
 	/* zero is "infinity" */
 	t->max_sectors = t->max_hw_sectors =
 		min_not_zero(t->max_sectors,b->max_sectors);
 	t->max_phys_segments = min(t->max_phys_segments,b->max_phys_segments);
 	t->max_hw_segments = min(t->max_hw_segments,b->max_hw_segments);
 	t->max_segment_size = min(t->max_segment_size,b->max_segment_size);
 	t->hardsect_size = max(t->hardsect_size,b->hardsect_size);
 }
 EXPORT_SYMBOL(blk_queue_stack_limits);
 /**
  * blk_queue_segment_boundary - set boundary rules for segment merging
  * @q:  the request queue for the device
  * @mask:  the memory boundary mask
  **/
 void blk_queue_segment_boundary(request_queue_t *q, unsigned long mask)
 {
 	if (mask < PAGE_CACHE_SIZE - 1) {
 		mask = PAGE_CACHE_SIZE - 1;
 		printk("%s: set to minimum %lx\n", __FUNCTION__, mask);
 	}
 	q->seg_boundary_mask = mask;
 }
 EXPORT_SYMBOL(blk_queue_segment_boundary);
 /**
  * blk_queue_dma_alignment - set dma length and memory alignment
  * @q:     the request queue for the device
  * @mask:  alignment mask
  *
  * description:
  *    set required memory and length aligment for direct dma transactions.
  *    this is used when buiding direct io requests for the queue.
  *
  **/
 void blk_queue_dma_alignment(request_queue_t *q, int mask)
 {
 	q->dma_alignment = mask;
 }
 EXPORT_SYMBOL(blk_queue_dma_alignment);
 /**
  * blk_queue_find_tag - find a request by its tag and queue
- *
  * @q:	 The request queue for the device
  * @tag: The tag of the request
  *
  * Notes:
  *    Should be used when a device returns a tag and you want to match
  *    it with a request.
  *
  *    no locks need be held.
  **/
 struct request *blk_queue_find_tag(request_queue_t *q, int tag)
 {
 	struct blk_queue_tag *bqt = q->queue_tags;
 	if (unlikely(bqt == NULL || tag >= bqt->real_max_depth))
 		return NULL;
 	return bqt->tag_index[tag];
 }
 EXPORT_SYMBOL(blk_queue_find_tag);
 /**
  * __blk_queue_free_tags - release tag maintenance info
  * @q:  the request queue for the device
  *
  *  Notes:
  *    blk_cleanup_queue() will take care of calling this function, if tagging
  *    has been used. So there's no need to call this directly.
  **/
 static void __blk_queue_free_tags(request_queue_t *q)
 {
 	struct blk_queue_tag *bqt = q->queue_tags;
 	if (!bqt)
 		return;
 	if (atomic_dec_and_test(&bqt->refcnt)) {
 		BUG_ON(bqt->busy);
 		BUG_ON(!list_empty(&bqt->busy_list));
 		kfree(bqt->tag_index);
 		bqt->tag_index = NULL;
 		kfree(bqt->tag_map);
 		bqt->tag_map = NULL;
 		kfree(bqt);
 	}
 	q->queue_tags = NULL;
 	q->queue_flags &= ~(1 << QUEUE_FLAG_QUEUED);
 }
 /**
  * blk_queue_free_tags - release tag maintenance info
  * @q:  the request queue for the device
  *
  *  Notes:
  *	This is used to disabled tagged queuing to a device, yet leave
  *	queue in function.
  **/
 void blk_queue_free_tags(request_queue_t *q)
 {
 	clear_bit(QUEUE_FLAG_QUEUED, &q->queue_flags);
 }
 EXPORT_SYMBOL(blk_queue_free_tags);
 static int
 init_tag_map(request_queue_t *q, struct blk_queue_tag *tags, int depth)
 {
 	struct request **tag_index;
 	unsigned long *tag_map;
 	int nr_ulongs;
 	if (depth > q->nr_requests * 2) {
 		depth = q->nr_requests * 2;
 		printk(KERN_ERR "%s: adjusted depth to %d\n",
 				__FUNCTION__, depth);
 	}
 	tag_index = kmalloc(depth * sizeof(struct request *), GFP_ATOMIC);
 	if (!tag_index)
 		goto fail;
 	nr_ulongs = ALIGN(depth, BITS_PER_LONG) / BITS_PER_LONG;
 	tag_map = kmalloc(nr_ulongs * sizeof(unsigned long), GFP_ATOMIC);
 	if (!tag_map)
 		goto fail;
 	memset(tag_index, 0, depth * sizeof(struct request *));
 	memset(tag_map, 0, nr_ulongs * sizeof(unsigned long));
 	tags->real_max_depth = depth;
 	tags->max_depth = depth;
 	tags->tag_index = tag_index;
 	tags->tag_map = tag_map;
 	return 0;
 fail:
 	kfree(tag_index);
 	return -ENOMEM;
 }
 /**
  * blk_queue_init_tags - initialize the queue tag info
  * @q:  the request queue for the device
  * @depth:  the maximum queue depth supported
  * @tags: the tag to use
  **/
 int blk_queue_init_tags(request_queue_t *q, int depth,
 			struct blk_queue_tag *tags)
 {
 	int rc;
 	BUG_ON(tags && q->queue_tags && tags != q->queue_tags);
 	if (!tags && !q->queue_tags) {
 		tags = kmalloc(sizeof(struct blk_queue_tag), GFP_ATOMIC);
 		if (!tags)
 			goto fail;
 		if (init_tag_map(q, tags, depth))
 			goto fail;
 		INIT_LIST_HEAD(&tags->busy_list);
 		tags->busy = 0;
 		atomic_set(&tags->refcnt, 1);
 	} else if (q->queue_tags) {
 		if ((rc = blk_queue_resize_tags(q, depth)))
 			return rc;
 		set_bit(QUEUE_FLAG_QUEUED, &q->queue_flags);
 		return 0;
 	} else
 		atomic_inc(&tags->refcnt);
 	/*
 	 * assign it, all done
 	 */
 	q->queue_tags = tags;
 	q->queue_flags |= (1 << QUEUE_FLAG_QUEUED);
 	return 0;
 fail:
 	kfree(tags);
 	return -ENOMEM;
 }
 EXPORT_SYMBOL(blk_queue_init_tags);
 /**
  * blk_queue_resize_tags - change the queueing depth
  * @q:  the request queue for the device
  * @new_depth: the new max command queueing depth
  *
  *  Notes:
  *    Must be called with the queue lock held.
  **/
 int blk_queue_resize_tags(request_queue_t *q, int new_depth)
 {
 	struct blk_queue_tag *bqt = q->queue_tags;
 	struct request **tag_index;
 	unsigned long *tag_map;
 	int max_depth, nr_ulongs;
 	if (!bqt)
 		return -ENXIO;
 	/*
 	 * if we already have large enough real_max_depth.  just
 	 * adjust max_depth.  *NOTE* as requests with tag value
 	 * between new_depth and real_max_depth can be in-flight, tag
 	 * map can not be shrunk blindly here.
 	 */
 	if (new_depth <= bqt->real_max_depth) {
 		bqt->max_depth = new_depth;
 		return 0;
 	}
 	/*
 	 * save the old state info, so we can copy it back
 	 */
 	tag_index = bqt->tag_index;
 	tag_map = bqt->tag_map;
 	max_depth = bqt->real_max_depth;
 	if (init_tag_map(q, bqt, new_depth))
 		return -ENOMEM;
 	memcpy(bqt->tag_index, tag_index, max_depth * sizeof(struct request *));
 	nr_ulongs = ALIGN(max_depth, BITS_PER_LONG) / BITS_PER_LONG;
 	memcpy(bqt->tag_map, tag_map, nr_ulongs * sizeof(unsigned long));
 	kfree(tag_index);
 	kfree(tag_map);
 	return 0;
 }
 EXPORT_SYMBOL(blk_queue_resize_tags);
 /**
  * blk_queue_end_tag - end tag operations for a request
  * @q:  the request queue for the device
  * @rq: the request that has completed
  *
  *  Description:
  *    Typically called when end_that_request_first() returns 0, meaning
  *    all transfers have been done for a request. It's important to call
  *    this function before end_that_request_last(), as that will put the
  *    request back on the free list thus corrupting the internal tag list.
  *
  *  Notes:
  *   queue lock must be held.
  **/
 void blk_queue_end_tag(request_queue_t *q, struct request *rq)
 {
 	struct blk_queue_tag *bqt = q->queue_tags;
 	int tag = rq->tag;
 	BUG_ON(tag == -1);
 	if (unlikely(tag >= bqt->real_max_depth))
 		/*
 		 * This can happen after tag depth has been reduced.
 		 * FIXME: how about a warning or info message here?
 		 */
 		return;
 	if (unlikely(!__test_and_clear_bit(tag, bqt->tag_map))) {
 		printk(KERN_ERR "%s: attempt to clear non-busy tag (%d)\n",
 		       __FUNCTION__, tag);
 		return;
 	}
 	list_del_init(&rq->queuelist);
 	rq->flags &= ~REQ_QUEUED;
 	rq->tag = -1;
 	if (unlikely(bqt->tag_index[tag] == NULL))
 		printk(KERN_ERR "%s: tag %d is missing\n",
 		       __FUNCTION__, tag);
 	bqt->tag_index[tag] = NULL;
 	bqt->busy--;
 }
 EXPORT_SYMBOL(blk_queue_end_tag);
 /**
  * blk_queue_start_tag - find a free tag and assign it
  * @q:  the request queue for the device
  * @rq:  the block request that needs tagging
  *
  *  Description:
  *    This can either be used as a stand-alone helper, or possibly be
  *    assigned as the queue &prep_rq_fn (in which case &struct request
  *    automagically gets a tag assigned). Note that this function
  *    assumes that any type of request can be queued! if this is not
  *    true for your device, you must check the request type before
  *    calling this function.  The request will also be removed from
  *    the request queue, so it's the drivers responsibility to readd
  *    it if it should need to be restarted for some reason.
  *
  *  Notes:
  *   queue lock must be held.
  **/
 int blk_queue_start_tag(request_queue_t *q, struct request *rq)
 {
 	struct blk_queue_tag *bqt = q->queue_tags;
 	int tag;
 	if (unlikely((rq->flags & REQ_QUEUED))) {
 		printk(KERN_ERR
 		       "%s: request %p for device [%s] already tagged %d",
 		       __FUNCTION__, rq,
 		       rq->rq_disk ? rq->rq_disk->disk_name : "?", rq->tag);
 		BUG();
 	}
 	tag = find_first_zero_bit(bqt->tag_map, bqt->max_depth);
 	if (tag >= bqt->max_depth)
 		return 1;
 	__set_bit(tag, bqt->tag_map);
 	rq->flags |= REQ_QUEUED;
 	rq->tag = tag;
 	bqt->tag_index[tag] = rq;
 	blkdev_dequeue_request(rq);
 	list_add(&rq->queuelist, &bqt->busy_list);
 	bqt->busy++;
 	return 0;
 }
 EXPORT_SYMBOL(blk_queue_start_tag);
 /**
  * blk_queue_invalidate_tags - invalidate all pending tags
  * @q:  the request queue for the device
  *
  *  Description:
  *   Hardware conditions may dictate a need to stop all pending requests.
  *   In this case, we will safely clear the block side of the tag queue and
  *   readd all requests to the request queue in the right order.
  *
  *  Notes:
  *   queue lock must be held.
  **/
 void blk_queue_invalidate_tags(request_queue_t *q)
 {
 	struct blk_queue_tag *bqt = q->queue_tags;
 	struct list_head *tmp, *n;
 	struct request *rq;
 	list_for_each_safe(tmp, n, &bqt->busy_list) {
 		rq = list_entry_rq(tmp);
 		if (rq->tag == -1) {
 			printk(KERN_ERR
 			       "%s: bad tag found on list\n", __FUNCTION__);
 			list_del_init(&rq->queuelist);
 			rq->flags &= ~REQ_QUEUED;
 		} else
 			blk_queue_end_tag(q, rq);
 		rq->flags &= ~REQ_STARTED;
 		__elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 0);
 	}
 }
 EXPORT_SYMBOL(blk_queue_invalidate_tags);
 static char *rq_flags[] = {
 	"REQ_RW",
 	"REQ_FAILFAST",
 	"REQ_SORTED",
 	"REQ_SOFTBARRIER",
 	"REQ_HARDBARRIER",
 	"REQ_CMD",
 	"REQ_NOMERGE",
 	"REQ_STARTED",
 	"REQ_DONTPREP",
 	"REQ_QUEUED",
 	"REQ_ELVPRIV",
 	"REQ_PC",
 	"REQ_BLOCK_PC",
 	"REQ_SENSE",
 	"REQ_FAILED",
 	"REQ_QUIET",
 	"REQ_SPECIAL",
 	"REQ_DRIVE_CMD",
 	"REQ_DRIVE_TASK",
 	"REQ_DRIVE_TASKFILE",
 	"REQ_PREEMPT",
 	"REQ_PM_SUSPEND",
 	"REQ_PM_RESUME",
 	"REQ_PM_SHUTDOWN",
 };
 void blk_dump_rq_flags(struct request *rq, char *msg)
 {
 	int bit;
 	printk("%s: dev %s: flags = ", msg,
 		rq->rq_disk ? rq->rq_disk->disk_name : "?");
 	bit = 0;
 	do {
 		if (rq->flags & (1 << bit))
 			printk("%s ", rq_flags[bit]);
 		bit++;
 	} while (bit < __REQ_NR_BITS);
 	printk("\nsector %llu, nr/cnr %lu/%u\n", (unsigned long long)rq->sector,
 						       rq->nr_sectors,
 						       rq->current_nr_sectors);
 	printk("bio %p, biotail %p, buffer %p, data %p, len %u\n", rq->bio, rq->biotail, rq->buffer, rq->data, rq->data_len);
 	if (rq->flags & (REQ_BLOCK_PC | REQ_PC)) {
 		printk("cdb: ");
 		for (bit = 0; bit < sizeof(rq->cmd); bit++)
 			printk("%02x ", rq->cmd[bit]);
 		printk("\n");
 	}
 }
 EXPORT_SYMBOL(blk_dump_rq_flags);
 void blk_recount_segments(request_queue_t *q, struct bio *bio)
 {
 	struct bio_vec *bv, *bvprv = NULL;
 	int i, nr_phys_segs, nr_hw_segs, seg_size, hw_seg_size, cluster;
 	int high, highprv = 1;
 	if (unlikely(!bio->bi_io_vec))
 		return;
 	cluster = q->queue_flags & (1 << QUEUE_FLAG_CLUSTER);
 	hw_seg_size = seg_size = nr_phys_segs = nr_hw_segs = 0;
 	bio_for_each_segment(bv, bio, i) {
 		/*
 		 * the trick here is making sure that a high page is never
 		 * considered part of another segment, since that might
 		 * change with the bounce page.
 		 */
 		high = page_to_pfn(bv->bv_page) >= q->bounce_pfn;
 		if (high || highprv)
 			goto new_hw_segment;
 		if (cluster) {
 			if (seg_size + bv->bv_len > q->max_segment_size)
 				goto new_segment;
 			if (!BIOVEC_PHYS_MERGEABLE(bvprv, bv))
 				goto new_segment;
 			if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bv))
 				goto new_segment;
 			if (BIOVEC_VIRT_OVERSIZE(hw_seg_size + bv->bv_len))
 				goto new_hw_segment;
 			seg_size += bv->bv_len;
 			hw_seg_size += bv->bv_len;
 			bvprv = bv;
 			continue;
 		}
 new_segment:
 		if (BIOVEC_VIRT_MERGEABLE(bvprv, bv) &&
 		    !BIOVEC_VIRT_OVERSIZE(hw_seg_size + bv->bv_len)) {
 			hw_seg_size += bv->bv_len;
 		} else {
 new_hw_segment:
 			if (hw_seg_size > bio->bi_hw_front_size)
 				bio->bi_hw_front_size = hw_seg_size;
 			hw_seg_size = BIOVEC_VIRT_START_SIZE(bv) + bv->bv_len;
 			nr_hw_segs++;
 		}
 		nr_phys_segs++;
 		bvprv = bv;
 		seg_size = bv->bv_len;
 		highprv = high;
 	}
 	if (hw_seg_size > bio->bi_hw_back_size)
 		bio->bi_hw_back_size = hw_seg_size;
 	if (nr_hw_segs == 1 && hw_seg_size > bio->bi_hw_front_size)
 		bio->bi_hw_front_size = hw_seg_size;
 	bio->bi_phys_segments = nr_phys_segs;
 	bio->bi_hw_segments = nr_hw_segs;
 	bio->bi_flags |= (1 << BIO_SEG_VALID);
 }
 static int blk_phys_contig_segment(request_queue_t *q, struct bio *bio,
 				   struct bio *nxt)
 {
 	if (!(q->queue_flags & (1 << QUEUE_FLAG_CLUSTER)))
 		return 0;
 	if (!BIOVEC_PHYS_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)))
 		return 0;
 	if (bio->bi_size + nxt->bi_size > q->max_segment_size)
 		return 0;
 	/*
 	 * bio and nxt are contigous in memory, check if the queue allows
 	 * these two to be merged into one
 	 */
 	if (BIO_SEG_BOUNDARY(q, bio, nxt))
 		return 1;
 	return 0;
 }
 static int blk_hw_contig_segment(request_queue_t *q, struct bio *bio,
 				 struct bio *nxt)
 {
 	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
 		blk_recount_segments(q, bio);
 	if (unlikely(!bio_flagged(nxt, BIO_SEG_VALID)))
 		blk_recount_segments(q, nxt);
 	if (!BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(nxt)) ||
 	    BIOVEC_VIRT_OVERSIZE(bio->bi_hw_front_size + bio->bi_hw_back_size))
 		return 0;
 	if (bio->bi_size + nxt->bi_size > q->max_segment_size)
 		return 0;
 	return 1;
 }
 /*
  * map a request to scatterlist, return number of sg entries setup. Caller
  * must make sure sg can hold rq->nr_phys_segments entries
  */
 int blk_rq_map_sg(request_queue_t *q, struct request *rq, struct scatterlist *sg)
 {
 	struct bio_vec *bvec, *bvprv;
 	struct bio *bio;
 	int nsegs, i, cluster;
 	nsegs = 0;
 	cluster = q->queue_flags & (1 << QUEUE_FLAG_CLUSTER);
 	/*
 	 * for each bio in rq
 	 */
 	bvprv = NULL;
 	rq_for_each_bio(bio, rq) {
 		/*
 		 * for each segment in bio
 		 */
 		bio_for_each_segment(bvec, bio, i) {
 			int nbytes = bvec->bv_len;
 			if (bvprv && cluster) {
 				if (sg[nsegs - 1].length + nbytes > q->max_segment_size)
 					goto new_segment;
 				if (!BIOVEC_PHYS_MERGEABLE(bvprv, bvec))
 					goto new_segment;
 				if (!BIOVEC_SEG_BOUNDARY(q, bvprv, bvec))
 					goto new_segment;
 				sg[nsegs - 1].length += nbytes;
 			} else {
 new_segment:
 				memset(&sg[nsegs],0,sizeof(struct scatterlist));
 				sg[nsegs].page = bvec->bv_page;
 				sg[nsegs].length = nbytes;
 				sg[nsegs].offset = bvec->bv_offset;
 				nsegs++;
 			}
 			bvprv = bvec;
 		} /* segments in bio */
 	} /* bios in rq */
 	return nsegs;
 }
 EXPORT_SYMBOL(blk_rq_map_sg);
 /*
  * the standard queue merge functions, can be overridden with device
  * specific ones if so desired
  */
 static inline int ll_new_mergeable(request_queue_t *q,
 				   struct request *req,
 				   struct bio *bio)
 {
 	int nr_phys_segs = bio_phys_segments(q, bio);
 	if (req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) {
 		req->flags |= REQ_NOMERGE;
 		if (req == q->last_merge)
 			q->last_merge = NULL;
 		return 0;
 	}
 	/*
 	 * A hw segment is just getting larger, bump just the phys
 	 * counter.
 	 */
 	req->nr_phys_segments += nr_phys_segs;
 	return 1;
 }
 static inline int ll_new_hw_segment(request_queue_t *q,
 				    struct request *req,
 				    struct bio *bio)
 {
 	int nr_hw_segs = bio_hw_segments(q, bio);
 	int nr_phys_segs = bio_phys_segments(q, bio);
 	if (req->nr_hw_segments + nr_hw_segs > q->max_hw_segments
 	    || req->nr_phys_segments + nr_phys_segs > q->max_phys_segments) {
 		req->flags |= REQ_NOMERGE;
 		if (req == q->last_merge)
 			q->last_merge = NULL;
 		return 0;
 	}
 	/*
 	 * This will form the start of a new hw segment.  Bump both
 	 * counters.
 	 */
 	req->nr_hw_segments += nr_hw_segs;
 	req->nr_phys_segments += nr_phys_segs;
 	return 1;
 }
 static int ll_back_merge_fn(request_queue_t *q, struct request *req,
 			    struct bio *bio)
 {
 	int len;
 	if (req->nr_sectors + bio_sectors(bio) > q->max_sectors) {
 		req->flags |= REQ_NOMERGE;
 		if (req == q->last_merge)
 			q->last_merge = NULL;
 		return 0;
 	}
 	if (unlikely(!bio_flagged(req->biotail, BIO_SEG_VALID)))
 		blk_recount_segments(q, req->biotail);
 	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
 		blk_recount_segments(q, bio);
 	len = req->biotail->bi_hw_back_size + bio->bi_hw_front_size;
 	if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(req->biotail), __BVEC_START(bio)) &&
 	    !BIOVEC_VIRT_OVERSIZE(len)) {
 		int mergeable =  ll_new_mergeable(q, req, bio);
 		if (mergeable) {
 			if (req->nr_hw_segments == 1)
 				req->bio->bi_hw_front_size = len;
 			if (bio->bi_hw_segments == 1)
 				bio->bi_hw_back_size = len;
 		}
 		return mergeable;
 	}
 	return ll_new_hw_segment(q, req, bio);
 }
 static int ll_front_merge_fn(request_queue_t *q, struct request *req,
 			     struct bio *bio)
 {
 	int len;
 	if (req->nr_sectors + bio_sectors(bio) > q->max_sectors) {
 		req->flags |= REQ_NOMERGE;
 		if (req == q->last_merge)
 			q->last_merge = NULL;
 		return 0;
 	}
 	len = bio->bi_hw_back_size + req->bio->bi_hw_front_size;
 	if (unlikely(!bio_flagged(bio, BIO_SEG_VALID)))
 		blk_recount_segments(q, bio);
 	if (unlikely(!bio_flagged(req->bio, BIO_SEG_VALID)))
 		blk_recount_segments(q, req->bio);
 	if (BIOVEC_VIRT_MERGEABLE(__BVEC_END(bio), __BVEC_START(req->bio)) &&
 	    !BIOVEC_VIRT_OVERSIZE(len)) {
 		int mergeable =  ll_new_mergeable(q, req, bio);
 		if (mergeable) {
 			if (bio->bi_hw_segments == 1)
 				bio->bi_hw_front_size = len;
 			if (req->nr_hw_segments == 1)
 				req->biotail->bi_hw_back_size = len;
 		}
 		return mergeable;
 	}
 	return ll_new_hw_segment(q, req, bio);
 }
 static int ll_merge_requests_fn(request_queue_t *q, struct request *req,
 				struct request *next)
 {
 	int total_phys_segments;
 	int total_hw_segments;
 	/*
 	 * First check if the either of the requests are re-queued
 	 * requests.  Can't merge them if they are.
 	 */
 	if (req->special || next->special)
 		return 0;
 	/*
 	 * Will it become too large?
 	 */
 	if ((req->nr_sectors + next->nr_sectors) > q->max_sectors)
 		return 0;
 	total_phys_segments = req->nr_phys_segments + next->nr_phys_segments;
 	if (blk_phys_contig_segment(q, req->biotail, next->bio))
 		total_phys_segments--;
 	if (total_phys_segments > q->max_phys_segments)
 		return 0;
 	total_hw_segments = req->nr_hw_segments + next->nr_hw_segments;
 	if (blk_hw_contig_segment(q, req->biotail, next->bio)) {
 		int len = req->biotail->bi_hw_back_size + next->bio->bi_hw_front_size;
 		/*
 		 * propagate the combined length to the end of the requests
 		 */
 		if (req->nr_hw_segments == 1)
 			req->bio->bi_hw_front_size = len;
 		if (next->nr_hw_segments == 1)
 			next->biotail->bi_hw_back_size = len;
 		total_hw_segments--;
 	}
 	if (total_hw_segments > q->max_hw_segments)
 		return 0;
 	/* Merge is OK... */
 	req->nr_phys_segments = total_phys_segments;
 	req->nr_hw_segments = total_hw_segments;
 	return 1;
 }
 /*
  * "plug" the device if there are no outstanding requests: this will
  * force the transfer to start only after we have put all the requests
  * on the list.
  *
  * This is called with interrupts off and no requests on the queue and
  * with the queue lock held.
  */
 void blk_plug_device(request_queue_t *q)
 {
 	WARN_ON(!irqs_disabled());
 	/*
 	 * don't plug a stopped queue, it must be paired with blk_start_queue()
 	 * which will restart the queueing
 	 */
 	if (test_bit(QUEUE_FLAG_STOPPED, &q->queue_flags))
 		return;
 	if (!test_and_set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags))
 		mod_timer(&q->unplug_timer, jiffies + q->unplug_delay);
 }
 EXPORT_SYMBOL(blk_plug_device);
 /*
  * remove the queue from the plugged list, if present. called with
  * queue lock held and interrupts disabled.
  */
 int blk_remove_plug(request_queue_t *q)
 {
 	WARN_ON(!irqs_disabled());
 	if (!test_and_clear_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags))
 		return 0;
 	del_timer(&q->unplug_timer);
 	return 1;
 }
 EXPORT_SYMBOL(blk_remove_plug);
 /*
  * remove the plug and let it rip..
  */
 void __generic_unplug_device(request_queue_t *q)
 {
 	if (unlikely(test_bit(QUEUE_FLAG_STOPPED, &q->queue_flags)))
 		return;
 	if (!blk_remove_plug(q))
 		return;
 	q->request_fn(q);
 }
 EXPORT_SYMBOL(__generic_unplug_device);
 /**
  * generic_unplug_device - fire a request queue
  * @q:    The &request_queue_t in question
  *
  * Description:
  *   Linux uses plugging to build bigger requests queues before letting
  *   the device have at them. If a queue is plugged, the I/O scheduler
  *   is still adding and merging requests on the queue. Once the queue
  *   gets unplugged, the request_fn defined for the queue is invoked and
  *   transfers started.
  **/
 void generic_unplug_device(request_queue_t *q)
 {
 	spin_lock_irq(q->queue_lock);
 	__generic_unplug_device(q);
 	spin_unlock_irq(q->queue_lock);
 }
 EXPORT_SYMBOL(generic_unplug_device);
 static void blk_backing_dev_unplug(struct backing_dev_info *bdi,
 				   struct page *page)
 {
 	request_queue_t *q = bdi->unplug_io_data;
 	/*
 	 * devices don't necessarily have an ->unplug_fn defined
 	 */
 	if (q->unplug_fn)
 		q->unplug_fn(q);
 }
 static void blk_unplug_work(void *data)
 {
 	request_queue_t *q = data;
 	q->unplug_fn(q);
 }
 static void blk_unplug_timeout(unsigned long data)
 {
 	request_queue_t *q = (request_queue_t *)data;
 	kblockd_schedule_work(&q->unplug_work);
 }
 /**
  * blk_start_queue - restart a previously stopped queue
  * @q:    The &request_queue_t in question
  *
  * Description:
  *   blk_start_queue() will clear the stop flag on the queue, and call
  *   the request_fn for the queue if it was in a stopped state when
  *   entered. Also see blk_stop_queue(). Queue lock must be held.
  **/
 void blk_start_queue(request_queue_t *q)
 {
 	clear_bit(QUEUE_FLAG_STOPPED, &q->queue_flags);
 	/*
 	 * one level of recursion is ok and is much faster than kicking
 	 * the unplug handling
 	 */
 	if (!test_and_set_bit(QUEUE_FLAG_REENTER, &q->queue_flags)) {
 		q->request_fn(q);
 		clear_bit(QUEUE_FLAG_REENTER, &q->queue_flags);
 	} else {
 		blk_plug_device(q);
 		kblockd_schedule_work(&q->unplug_work);
 	}
 }
 EXPORT_SYMBOL(blk_start_queue);
 /**
  * blk_stop_queue - stop a queue
  * @q:    The &request_queue_t in question
  *
  * Description:
  *   The Linux block layer assumes that a block driver will consume all
  *   entries on the request queue when the request_fn strategy is called.
  *   Often this will not happen, because of hardware limitations (queue
  *   depth settings). If a device driver gets a 'queue full' response,
  *   or if it simply chooses not to queue more I/O at one point, it can
  *   call this function to prevent the request_fn from being called until
  *   the driver has signalled it's ready to go again. This happens by calling
  *   blk_start_queue() to restart queue operations. Queue lock must be held.
  **/
 void blk_stop_queue(request_queue_t *q)
 {
 	blk_remove_plug(q);
 	set_bit(QUEUE_FLAG_STOPPED, &q->queue_flags);
 }
 EXPORT_SYMBOL(blk_stop_queue);
 /**
  * blk_sync_queue - cancel any pending callbacks on a queue
  * @q: the queue
  *
  * Description:
  *     The block layer may perform asynchronous callback activity
  *     on a queue, such as calling the unplug function after a timeout.
  *     A block device may call blk_sync_queue to ensure that any
  *     such activity is cancelled, thus allowing it to release resources
  *     the the callbacks might use. The caller must already have made sure
  *     that its ->make_request_fn will not re-add plugging prior to calling
  *     this function.
  *
  */
 void blk_sync_queue(struct request_queue *q)
 {
 	del_timer_sync(&q->unplug_timer);
 	kblockd_flush();
 }
 EXPORT_SYMBOL(blk_sync_queue);
 /**
  * blk_run_queue - run a single device queue
  * @q:	The queue to run
  */
 void blk_run_queue(struct request_queue *q)
 {
 	unsigned long flags;
 	spin_lock_irqsave(q->queue_lock, flags);
 	blk_remove_plug(q);
 	if (!elv_queue_empty(q))
 		q->request_fn(q);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 EXPORT_SYMBOL(blk_run_queue);
 /**
  * blk_cleanup_queue: - release a &request_queue_t when it is no longer needed
  * @q:    the request queue to be released
  *
  * Description:
  *     blk_cleanup_queue is the pair to blk_init_queue() or
  *     blk_queue_make_request().  It should be called when a request queue is
  *     being released; typically when a block device is being de-registered.
  *     Currently, its primary task it to free all the &struct request
  *     structures that were allocated to the queue and the queue itself.
  *
  * Caveat:
  *     Hopefully the low level driver will have finished any
  *     outstanding requests first...
  **/
 void blk_cleanup_queue(request_queue_t * q)
 {
 	struct request_list *rl = &q->rq;
 	if (!atomic_dec_and_test(&q->refcnt))
 		return;
 	if (q->elevator)
 		elevator_exit(q->elevator);
 	blk_sync_queue(q);
 	if (rl->rq_pool)
 		mempool_destroy(rl->rq_pool);
 	if (q->queue_tags)
 		__blk_queue_free_tags(q);
 	blk_queue_ordered(q, QUEUE_ORDERED_NONE);
 	kmem_cache_free(requestq_cachep, q);
 }
 EXPORT_SYMBOL(blk_cleanup_queue);
 static int blk_init_free_list(request_queue_t *q)
 {
 	struct request_list *rl = &q->rq;
 	rl->count[READ] = rl->count[WRITE] = 0;
 	rl->starved[READ] = rl->starved[WRITE] = 0;
 	rl->elvpriv = 0;
 	init_waitqueue_head(&rl->wait[READ]);
 	init_waitqueue_head(&rl->wait[WRITE]);
 	rl->rq_pool = mempool_create_node(BLKDEV_MIN_RQ, mempool_alloc_slab,
 				mempool_free_slab, request_cachep, q->node);
 	if (!rl->rq_pool)
 		return -ENOMEM;
 	return 0;
 }
 static int __make_request(request_queue_t *, struct bio *);
 request_queue_t *blk_alloc_queue(gfp_t gfp_mask)
 {
 	return blk_alloc_queue_node(gfp_mask, -1);
 }
 EXPORT_SYMBOL(blk_alloc_queue);
 request_queue_t *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 {
 	request_queue_t *q;
 	q = kmem_cache_alloc_node(requestq_cachep, gfp_mask, node_id);
 	if (!q)
 		return NULL;
 	memset(q, 0, sizeof(*q));
 	init_timer(&q->unplug_timer);
 	atomic_set(&q->refcnt, 1);
 	q->backing_dev_info.unplug_io_fn = blk_backing_dev_unplug;
 	q->backing_dev_info.unplug_io_data = q;
 	return q;
 }
 EXPORT_SYMBOL(blk_alloc_queue_node);
 /**
  * blk_init_queue  - prepare a request queue for use with a block device
  * @rfn:  The function to be called to process requests that have been
  *        placed on the queue.
  * @lock: Request queue spin lock
  *
  * Description:
  *    If a block device wishes to use the standard request handling procedures,
  *    which sorts requests and coalesces adjacent requests, then it must
  *    call blk_init_queue().  The function @rfn will be called when there
  *    are requests on the queue that need to be processed.  If the device
  *    supports plugging, then @rfn may not be called immediately when requests
  *    are available on the queue, but may be called at some time later instead.
  *    Plugged queues are generally unplugged when a buffer belonging to one
  *    of the requests on the queue is needed, or due to memory pressure.
  *
  *    @rfn is not required, or even expected, to remove all requests off the
  *    queue, but only as many as it can handle at a time.  If it does leave
  *    requests on the queue, it is responsible for arranging that the requests
  *    get dealt with eventually.
  *
  *    The queue spin lock must be held while manipulating the requests on the
  *    request queue.
  *
  *    Function returns a pointer to the initialized request queue, or NULL if
  *    it didn't succeed.
  *
  * Note:
  *    blk_init_queue() must be paired with a blk_cleanup_queue() call
  *    when the block device is deactivated (such as at module unload).
  **/
 request_queue_t *blk_init_queue(request_fn_proc *rfn, spinlock_t *lock)
 {
 	return blk_init_queue_node(rfn, lock, -1);
 }
 EXPORT_SYMBOL(blk_init_queue);
 request_queue_t *
 blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id)
 {
 	request_queue_t *q = blk_alloc_queue_node(GFP_KERNEL, node_id);
 	if (!q)
 		return NULL;
 	q->node = node_id;
 	if (blk_init_free_list(q))
 		goto out_init;
 	/*
 	 * if caller didn't supply a lock, they get per-queue locking with
 	 * our embedded lock
 	 */
 	if (!lock) {
 		spin_lock_init(&q->__queue_lock);
 		lock = &q->__queue_lock;
 	}
 	q->request_fn		= rfn;
 	q->back_merge_fn       	= ll_back_merge_fn;
 	q->front_merge_fn      	= ll_front_merge_fn;
 	q->merge_requests_fn	= ll_merge_requests_fn;
 	q->prep_rq_fn		= NULL;
 	q->unplug_fn		= generic_unplug_device;
 	q->queue_flags		= (1 << QUEUE_FLAG_CLUSTER);
 	q->queue_lock		= lock;
 	blk_queue_segment_boundary(q, 0xffffffff);
 	blk_queue_make_request(q, __make_request);
 	blk_queue_max_segment_size(q, MAX_SEGMENT_SIZE);
 	blk_queue_max_hw_segments(q, MAX_HW_SEGMENTS);
 	blk_queue_max_phys_segments(q, MAX_PHYS_SEGMENTS);
 	/*
 	 * all done
 	 */
 	if (!elevator_init(q, NULL)) {
 		blk_queue_congestion_threshold(q);
 		return q;
 	}
 	blk_cleanup_queue(q);
 out_init:
 	kmem_cache_free(requestq_cachep, q);
 	return NULL;
 }
 EXPORT_SYMBOL(blk_init_queue_node);
 int blk_get_queue(request_queue_t *q)
 {
 	if (likely(!test_bit(QUEUE_FLAG_DEAD, &q->queue_flags))) {
 		atomic_inc(&q->refcnt);
 		return 0;
 	}
 	return 1;
 }
 EXPORT_SYMBOL(blk_get_queue);
 static inline void blk_free_request(request_queue_t *q, struct request *rq)
 {
 	if (rq->flags & REQ_ELVPRIV)
 		elv_put_request(q, rq);
 	mempool_free(rq, q->rq.rq_pool);
 }
 static inline struct request *
 blk_alloc_request(request_queue_t *q, int rw, struct bio *bio,
 		  int priv, gfp_t gfp_mask)
 {
 	struct request *rq = mempool_alloc(q->rq.rq_pool, gfp_mask);
 	if (!rq)
 		return NULL;
 	/*
 	 * first three bits are identical in rq->flags and bio->bi_rw,
 	 * see bio.h and blkdev.h
 	 */
 	rq->flags = rw;
 	if (priv) {
 		if (unlikely(elv_set_request(q, rq, bio, gfp_mask))) {
 			mempool_free(rq, q->rq.rq_pool);
 			return NULL;
 		}
 		rq->flags |= REQ_ELVPRIV;
 	}
 	return rq;
 }
 /*
  * ioc_batching returns true if the ioc is a valid batching request and
  * should be given priority access to a request.
  */
 static inline int ioc_batching(request_queue_t *q, struct io_context *ioc)
 {
 	if (!ioc)
 		return 0;
 	/*
 	 * Make sure the process is able to allocate at least 1 request
 	 * even if the batch times out, otherwise we could theoretically
 	 * lose wakeups.
 	 */
 	return ioc->nr_batch_requests == q->nr_batching ||
 		(ioc->nr_batch_requests > 0
 		&& time_before(jiffies, ioc->last_waited + BLK_BATCH_TIME));
 }
 /*
  * ioc_set_batching sets ioc to be a new "batcher" if it is not one. This
  * will cause the process to be a "batcher" on all queues in the system. This
  * is the behaviour we want though - once it gets a wakeup it should be given
  * a nice run.
  */
 static void ioc_set_batching(request_queue_t *q, struct io_context *ioc)
 {
 	if (!ioc || ioc_batching(q, ioc))
 		return;
 	ioc->nr_batch_requests = q->nr_batching;
 	ioc->last_waited = jiffies;
 }
 static void __freed_request(request_queue_t *q, int rw)
 {
 	struct request_list *rl = &q->rq;
 	if (rl->count[rw] < queue_congestion_off_threshold(q))
 		clear_queue_congested(q, rw);
 	if (rl->count[rw] + 1 <= q->nr_requests) {
 		if (waitqueue_active(&rl->wait[rw]))
 			wake_up(&rl->wait[rw]);
 		blk_clear_queue_full(q, rw);
 	}
 }
 /*
  * A request has just been released.  Account for it, update the full and
  * congestion status, wake up any waiters.   Called under q->queue_lock.
  */
 static void freed_request(request_queue_t *q, int rw, int priv)
 {
 	struct request_list *rl = &q->rq;
 	rl->count[rw]--;
 	if (priv)
 		rl->elvpriv--;
 	__freed_request(q, rw);
 	if (unlikely(rl->starved[rw ^ 1]))
 		__freed_request(q, rw ^ 1);
 }
 #define blkdev_free_rq(list) list_entry((list)->next, struct request, queuelist)
 /*
  * Get a free request, queue_lock must be held.
  * Returns NULL on failure, with queue_lock held.
  * Returns !NULL on success, with queue_lock *not held*.
  */
 static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
 				   gfp_t gfp_mask)
 {
 	struct request *rq = NULL;
 	struct request_list *rl = &q->rq;
 	struct io_context *ioc = current_io_context(GFP_ATOMIC);
 	int priv;
 	if (rl->count[rw]+1 >= q->nr_requests) {
 		/*
 		 * The queue will fill after this allocation, so set it as
 		 * full, and mark this process as "batching". This process
 		 * will be allowed to complete a batch of requests, others
 		 * will be blocked.
 		 */
 		if (!blk_queue_full(q, rw)) {
 			ioc_set_batching(q, ioc);
 			blk_set_queue_full(q, rw);
 		}
 	}
 	switch (elv_may_queue(q, rw, bio)) {
 		case ELV_MQUEUE_NO:
 			goto rq_starved;
 		case ELV_MQUEUE_MAY:
 			break;
 		case ELV_MQUEUE_MUST:
 			goto get_rq;
 	}
 	if (blk_queue_full(q, rw) && !ioc_batching(q, ioc)) {
 		/*
 		 * The queue is full and the allocating process is not a
 		 * "batcher", and not exempted by the IO scheduler
 		 */
 		goto out;
 	}
 get_rq:
 	/*
 	 * Only allow batching queuers to allocate up to 50% over the defined
 	 * limit of requests, otherwise we could have thousands of requests
 	 * allocated with any setting of ->nr_requests
 	 */
 	if (rl->count[rw] >= (3 * q->nr_requests / 2))
 		goto out;
 	rl->count[rw]++;
 	rl->starved[rw] = 0;
 	if (rl->count[rw] >= queue_congestion_on_threshold(q))
 		set_queue_congested(q, rw);
 	priv = !test_bit(QUEUE_FLAG_ELVSWITCH, &q->queue_flags);
 	if (priv)
 		rl->elvpriv++;
 	spin_unlock_irq(q->queue_lock);
 	rq = blk_alloc_request(q, rw, bio, priv, gfp_mask);
 	if (!rq) {
 		/*
 		 * Allocation failed presumably due to memory. Undo anything
 		 * we might have messed up.
 		 *
 		 * Allocating task should really be put onto the front of the
 		 * wait queue, but this is pretty rare.
 		 */
 		spin_lock_irq(q->queue_lock);
 		freed_request(q, rw, priv);
 		/*
 		 * in the very unlikely event that allocation failed and no
 		 * requests for this direction was pending, mark us starved
 		 * so that freeing of a request in the other direction will
 		 * notice us. another possible fix would be to split the
 		 * rq mempool into READ and WRITE
 		 */
 rq_starved:
 		if (unlikely(rl->count[rw] == 0))
 			rl->starved[rw] = 1;
 		goto out;
 	}
 	if (ioc_batching(q, ioc))
 		ioc->nr_batch_requests--;
 	rq_init(q, rq);
 	rq->rl = rl;
 out:
 	return rq;
 }
 /*
  * No available requests for this queue, unplug the device and wait for some
  * requests to become available.
  *
  * Called with q->queue_lock held, and returns with it unlocked.
  */
 static struct request *get_request_wait(request_queue_t *q, int rw,
 					struct bio *bio)
 {
 	struct request *rq;
 	rq = get_request(q, rw, bio, GFP_NOIO);
 	while (!rq) {
 		DEFINE_WAIT(wait);
 		struct request_list *rl = &q->rq;
 		prepare_to_wait_exclusive(&rl->wait[rw], &wait,
 				TASK_UNINTERRUPTIBLE);
 		rq = get_request(q, rw, bio, GFP_NOIO);
 		if (!rq) {
 			struct io_context *ioc;
 			__generic_unplug_device(q);
 			spin_unlock_irq(q->queue_lock);
 			io_schedule();
 			/*
 			 * After sleeping, we become a "batching" process and
 			 * will be able to allocate at least one request, and
 			 * up to a big batch of them for a small period time.
 			 * See ioc_batching, ioc_set_batching
 			 */
 			ioc = current_io_context(GFP_NOIO);
 			ioc_set_batching(q, ioc);
 			spin_lock_irq(q->queue_lock);
 		}
 		finish_wait(&rl->wait[rw], &wait);
 	}
 	return rq;
 }
 struct request *blk_get_request(request_queue_t *q, int rw, gfp_t gfp_mask)
 {
 	struct request *rq;
 	BUG_ON(rw != READ && rw != WRITE);
 	spin_lock_irq(q->queue_lock);
 	if (gfp_mask & __GFP_WAIT) {
 		rq = get_request_wait(q, rw, NULL);
 	} else {
 		rq = get_request(q, rw, NULL, gfp_mask);
 		if (!rq)
 			spin_unlock_irq(q->queue_lock);
 	}
 	/* q->queue_lock is unlocked at this point */
 	return rq;
 }
 EXPORT_SYMBOL(blk_get_request);
 /**
  * blk_requeue_request - put a request back on queue
  * @q:		request queue where request should be inserted
  * @rq:		request to be inserted
  *
  * Description:
  *    Drivers often keep queueing requests until the hardware cannot accept
  *    more, when that condition happens we need to put the request back
  *    on the queue. Must be called with queue lock held.
  */
 void blk_requeue_request(request_queue_t *q, struct request *rq)
 {
 	if (blk_rq_tagged(rq))
 		blk_queue_end_tag(q, rq);
 	elv_requeue_request(q, rq);
 }
 EXPORT_SYMBOL(blk_requeue_request);
 /**
  * blk_insert_request - insert a special request in to a request queue
  * @q:		request queue where request should be inserted
  * @rq:		request to be inserted
  * @at_head:	insert request at head or tail of queue
  * @data:	private data
  *
  * Description:
  *    Many block devices need to execute commands asynchronously, so they don't
  *    block the whole kernel from preemption during request execution.  This is
  *    accomplished normally by inserting aritficial requests tagged as
  *    REQ_SPECIAL in to the corresponding request queue, and letting them be
  *    scheduled for actual execution by the request queue.
  *
  *    We have the option of inserting the head or the tail of the queue.
  *    Typically we use the tail for new ioctls and so forth.  We use the head
  *    of the queue for things like a QUEUE_FULL message from a device, or a
  *    host that is unable to accept a particular command.
  */
 void blk_insert_request(request_queue_t *q, struct request *rq,
 			int at_head, void *data)
 {
 	int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK;
 	unsigned long flags;
 	/*
 	 * tell I/O scheduler that this isn't a regular read/write (ie it
 	 * must not attempt merges on this) and that it acts as a soft
 	 * barrier
 	 */
 	rq->flags |= REQ_SPECIAL | REQ_SOFTBARRIER;
 	rq->special = data;
 	spin_lock_irqsave(q->queue_lock, flags);
 	/*
 	 * If command is tagged, release the tag
 	 */
 	if (blk_rq_tagged(rq))
 		blk_queue_end_tag(q, rq);
 	drive_stat_acct(rq, rq->nr_sectors, 1);
 	__elv_add_request(q, rq, where, 0);
 	if (blk_queue_plugged(q))
 		__generic_unplug_device(q);
 	else
 		q->request_fn(q);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 EXPORT_SYMBOL(blk_insert_request);
 /**
  * blk_rq_map_user - map user data to a request, for REQ_BLOCK_PC usage
  * @q:		request queue where request should be inserted
  * @rq:		request structure to fill
  * @ubuf:	the user buffer
  * @len:	length of user data
  *
  * Description:
  *    Data will be mapped directly for zero copy io, if possible. Otherwise
  *    a kernel bounce buffer is used.
  *
  *    A matching blk_rq_unmap_user() must be issued at the end of io, while
  *    still in process context.
  *
  *    Note: The mapped bio may need to be bounced through blk_queue_bounce()
  *    before being submitted to the device, as pages mapped may be out of
  *    reach. It's the callers responsibility to make sure this happens. The
  *    original bio must be passed back in to blk_rq_unmap_user() for proper
  *    unmapping.
  */
 int blk_rq_map_user(request_queue_t *q, struct request *rq, void __user *ubuf,
 		    unsigned int len)
 {
 	unsigned long uaddr;
 	struct bio *bio;
 	int reading;
 	if (len > (q->max_sectors << 9))
 		return -EINVAL;
 	if (!len || !ubuf)
 		return -EINVAL;
 	reading = rq_data_dir(rq) == READ;
 	/*
 	 * if alignment requirement is satisfied, map in user pages for
 	 * direct dma. else, set up kernel bounce buffers
 	 */
 	uaddr = (unsigned long) ubuf;
 	if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q)))
 		bio = bio_map_user(q, NULL, uaddr, len, reading);
 	else
 		bio = bio_copy_user(q, uaddr, len, reading);
 	if (!IS_ERR(bio)) {
 		rq->bio = rq->biotail = bio;
 		blk_rq_bio_prep(q, rq, bio);
 		rq->buffer = rq->data = NULL;
 		rq->data_len = len;
 		return 0;
 	}
 	/*
 	 * bio is the err-ptr
 	 */
 	return PTR_ERR(bio);
 }
 EXPORT_SYMBOL(blk_rq_map_user);
 /**
  * blk_rq_map_user_iov - map user data to a request, for REQ_BLOCK_PC usage
  * @q:		request queue where request should be inserted
  * @rq:		request to map data to
  * @iov:	pointer to the iovec
  * @iov_count:	number of elements in the iovec
  *
  * Description:
  *    Data will be mapped directly for zero copy io, if possible. Otherwise
  *    a kernel bounce buffer is used.
  *
  *    A matching blk_rq_unmap_user() must be issued at the end of io, while
  *    still in process context.
  *
  *    Note: The mapped bio may need to be bounced through blk_queue_bounce()
  *    before being submitted to the device, as pages mapped may be out of
  *    reach. It's the callers responsibility to make sure this happens. The
  *    original bio must be passed back in to blk_rq_unmap_user() for proper
  *    unmapping.
  */
 int blk_rq_map_user_iov(request_queue_t *q, struct request *rq,
 			struct sg_iovec *iov, int iov_count)
 {
 	struct bio *bio;
 	if (!iov || iov_count <= 0)
 		return -EINVAL;
 	/* we don't allow misaligned data like bio_map_user() does.  If the
 	 * user is using sg, they're expected to know the alignment constraints
 	 * and respect them accordingly */
 	bio = bio_map_user_iov(q, NULL, iov, iov_count, rq_data_dir(rq)== READ);
 	if (IS_ERR(bio))
 		return PTR_ERR(bio);
 	rq->bio = rq->biotail = bio;
 	blk_rq_bio_prep(q, rq, bio);
 	rq->buffer = rq->data = NULL;
 	rq->data_len = bio->bi_size;
 	return 0;
 }
 EXPORT_SYMBOL(blk_rq_map_user_iov);
 /**
  * blk_rq_unmap_user - unmap a request with user data
  * @bio:	bio to be unmapped
  * @ulen:	length of user buffer
  *
  * Description:
  *    Unmap a bio previously mapped by blk_rq_map_user().
  */
 int blk_rq_unmap_user(struct bio *bio, unsigned int ulen)
 {
 	int ret = 0;
 	if (bio) {
 		if (bio_flagged(bio, BIO_USER_MAPPED))
 			bio_unmap_user(bio);
 		else
 			ret = bio_uncopy_user(bio);
 	}
 	return 0;
 }
 EXPORT_SYMBOL(blk_rq_unmap_user);
 /**
  * blk_rq_map_kern - map kernel data to a request, for REQ_BLOCK_PC usage
  * @q:		request queue where request should be inserted
  * @rq:		request to fill
  * @kbuf:	the kernel buffer
  * @len:	length of user data
  * @gfp_mask:	memory allocation flags
  */
 int blk_rq_map_kern(request_queue_t *q, struct request *rq, void *kbuf,
 		    unsigned int len, gfp_t gfp_mask)
 {
 	struct bio *bio;
 	if (len > (q->max_sectors << 9))
 		return -EINVAL;
 	if (!len || !kbuf)
 		return -EINVAL;
 	bio = bio_map_kern(q, kbuf, len, gfp_mask);
 	if (IS_ERR(bio))
 		return PTR_ERR(bio);
 	if (rq_data_dir(rq) == WRITE)
 		bio->bi_rw |= (1 << BIO_RW);
 	rq->bio = rq->biotail = bio;
 	blk_rq_bio_prep(q, rq, bio);
 	rq->buffer = rq->data = NULL;
 	rq->data_len = len;
 	return 0;
 }
 EXPORT_SYMBOL(blk_rq_map_kern);
 /**
  * blk_execute_rq_nowait - insert a request into queue for execution
  * @q:		queue to insert the request in
  * @bd_disk:	matching gendisk
  * @rq:		request to insert
  * @at_head:    insert request at head or tail of queue
  * @done:	I/O completion handler
  *
  * Description:
  *    Insert a fully prepared request at the back of the io scheduler queue
  *    for execution.  Don't wait for completion.
  */
 void blk_execute_rq_nowait(request_queue_t *q, struct gendisk *bd_disk,
 			   struct request *rq, int at_head,
 			   void (*done)(struct request *))
 {
 	int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK;
 	rq->rq_disk = bd_disk;
 	rq->flags |= REQ_NOMERGE;
 	rq->end_io = done;
 	elv_add_request(q, rq, where, 1);
 	generic_unplug_device(q);
 }
 /**
  * blk_execute_rq - insert a request into queue for execution
  * @q:		queue to insert the request in
  * @bd_disk:	matching gendisk
  * @rq:		request to insert
  * @at_head:    insert request at head or tail of queue
  *
  * Description:
  *    Insert a fully prepared request at the back of the io scheduler queue
  *    for execution and wait for completion.
  */
 int blk_execute_rq(request_queue_t *q, struct gendisk *bd_disk,
 		   struct request *rq, int at_head)
 {
 	DECLARE_COMPLETION(wait);
 	char sense[SCSI_SENSE_BUFFERSIZE];
 	int err = 0;
 	/*
 	 * we need an extra reference to the request, so we can look at
 	 * it after io completion
 	 */
 	rq->ref_count++;
 	if (!rq->sense) {
 		memset(sense, 0, sizeof(sense));
 		rq->sense = sense;
 		rq->sense_len = 0;
 	}
 	rq->waiting = &wait;
 	blk_execute_rq_nowait(q, bd_disk, rq, at_head, blk_end_sync_rq);
 	wait_for_completion(&wait);
 	rq->waiting = NULL;
 	if (rq->errors)
 		err = -EIO;
 	return err;
 }
 EXPORT_SYMBOL(blk_execute_rq);
 /**
  * blkdev_issue_flush - queue a flush
  * @bdev:	blockdev to issue flush for
  * @error_sector:	error sector
  *
  * Description:
  *    Issue a flush for the block device in question. Caller can supply
  *    room for storing the error offset in case of a flush error, if they
  *    wish to.  Caller must run wait_for_completion() on its own.
  */
 int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
 {
 	request_queue_t *q;
 	if (bdev->bd_disk == NULL)
 		return -ENXIO;
 	q = bdev_get_queue(bdev);
 	if (!q)
 		return -ENXIO;
 	if (!q->issue_flush_fn)
 		return -EOPNOTSUPP;
 	return q->issue_flush_fn(q, bdev->bd_disk, error_sector);
 }
 EXPORT_SYMBOL(blkdev_issue_flush);
 static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io)
 {
 	int rw = rq_data_dir(rq);
 	if (!blk_fs_request(rq) || !rq->rq_disk)
 		return;
 	if (!new_io) {
 		__disk_stat_inc(rq->rq_disk, merges[rw]);
 	} else {
 		disk_round_stats(rq->rq_disk);
 		rq->rq_disk->in_flight++;
 	}
 }
 /*
  * add-request adds a request to the linked list.
  * queue lock is held and interrupts disabled, as we muck with the
  * request queue list.
  */
 static inline void add_request(request_queue_t * q, struct request * req)
 {
 	drive_stat_acct(req, req->nr_sectors, 1);
 	if (q->activity_fn)
 		q->activity_fn(q->activity_data, rq_data_dir(req));
 	/*
 	 * elevator indicated where it wants this request to be
 	 * inserted at elevator_merge time
 	 */
 	__elv_add_request(q, req, ELEVATOR_INSERT_SORT, 0);
 }
 /*
  * disk_round_stats()	- Round off the performance stats on a struct
  * disk_stats.
  *
  * The average IO queue length and utilisation statistics are maintained
  * by observing the current state of the queue length and the amount of
  * time it has been in this state for.
  *
  * Normally, that accounting is done on IO completion, but that can result
  * in more than a second's worth of IO being accounted for within any one
  * second, leading to >100% utilisation.  To deal with that, we call this
  * function to do a round-off before returning the results when reading
  * /proc/diskstats.  This accounts immediately for all queue usage up to
  * the current jiffies and restarts the counters again.
  */
 void disk_round_stats(struct gendisk *disk)
 {
 	unsigned long now = jiffies;
 	if (now == disk->stamp)
 		return;
 	if (disk->in_flight) {
 		__disk_stat_add(disk, time_in_queue,
 				disk->in_flight * (now - disk->stamp));
 		__disk_stat_add(disk, io_ticks, (now - disk->stamp));
 	}
 	disk->stamp = now;
 }
 /*
  * queue lock must be held
  */
 static void __blk_put_request(request_queue_t *q, struct request *req)
 {
 	struct request_list *rl = req->rl;
 	if (unlikely(!q))
 		return;
 	if (unlikely(--req->ref_count))
 		return;
 	elv_completed_request(q, req);
 	req->rq_status = RQ_INACTIVE;
 	req->rl = NULL;
 	/*
 	 * Request may not have originated from ll_rw_blk. if not,
 	 * it didn't come out of our reserved rq pools
 	 */
 	if (rl) {
 		int rw = rq_data_dir(req);
 		int priv = req->flags & REQ_ELVPRIV;
 		BUG_ON(!list_empty(&req->queuelist));
 		blk_free_request(q, req);
 		freed_request(q, rw, priv);
 	}
 }
 void blk_put_request(struct request *req)
 {
 	unsigned long flags;
 	request_queue_t *q = req->q;
 	/*
 	 * Gee, IDE calls in w/ NULL q.  Fix IDE and remove the
 	 * following if (q) test.
 	 */
 	if (q) {
 		spin_lock_irqsave(q->queue_lock, flags);
 		__blk_put_request(q, req);
 		spin_unlock_irqrestore(q->queue_lock, flags);
 	}
 }
 EXPORT_SYMBOL(blk_put_request);
 /**
  * blk_end_sync_rq - executes a completion event on a request
  * @rq: request to complete
  */
 void blk_end_sync_rq(struct request *rq)
 {
 	struct completion *waiting = rq->waiting;
 	rq->waiting = NULL;
 	__blk_put_request(rq->q, rq);
 	/*
 	 * complete last, if this is a stack request the process (and thus
 	 * the rq pointer) could be invalid right after this complete()
 	 */
 	complete(waiting);
 }
 EXPORT_SYMBOL(blk_end_sync_rq);
 /**
  * blk_congestion_wait - wait for a queue to become uncongested
  * @rw: READ or WRITE
  * @timeout: timeout in jiffies
  *
  * Waits for up to @timeout jiffies for a queue (any queue) to exit congestion.
  * If no queues are congested then just wait for the next request to be
  * returned.
  */
 long blk_congestion_wait(int rw, long timeout)
 {
 	long ret;
 	DEFINE_WAIT(wait);
 	wait_queue_head_t *wqh = &congestion_wqh[rw];
 	prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
 	ret = io_schedule_timeout(timeout);
 	finish_wait(wqh, &wait);
 	return ret;
 }
 EXPORT_SYMBOL(blk_congestion_wait);
 /*
  * Has to be called with the request spinlock acquired
  */
 static int attempt_merge(request_queue_t *q, struct request *req,
 			  struct request *next)
 {
 	if (!rq_mergeable(req) || !rq_mergeable(next))
 		return 0;
 	/*
 	 * not contigious
 	 */
 	if (req->sector + req->nr_sectors != next->sector)
 		return 0;
 	if (rq_data_dir(req) != rq_data_dir(next)
 	    || req->rq_disk != next->rq_disk
 	    || next->waiting || next->special)
 		return 0;
 	/*
 	 * If we are allowed to merge, then append bio list
 	 * from next to rq and release next. merge_requests_fn
 	 * will have updated segment counts, update sector
 	 * counts here.
 	 */
 	if (!q->merge_requests_fn(q, req, next))
 		return 0;
 	/*
 	 * At this point we have either done a back merge
 	 * or front merge. We need the smaller start_time of
 	 * the merged requests to be the current request
 	 * for accounting purposes.
 	 */
 	if (time_after(req->start_time, next->start_time))
 		req->start_time = next->start_time;
 	req->biotail->bi_next = next->bio;
 	req->biotail = next->biotail;
 	req->nr_sectors = req->hard_nr_sectors += next->hard_nr_sectors;
 	elv_merge_requests(q, req, next);
 	if (req->rq_disk) {
 		disk_round_stats(req->rq_disk);
 		req->rq_disk->in_flight--;
 	}
 	req->ioprio = ioprio_best(req->ioprio, next->ioprio);
 	__blk_put_request(q, next);
 	return 1;
 }
 static inline int attempt_back_merge(request_queue_t *q, struct request *rq)
 {
 	struct request *next = elv_latter_request(q, rq);
 	if (next)
 		return attempt_merge(q, rq, next);
 	return 0;
 }
 static inline int attempt_front_merge(request_queue_t *q, struct request *rq)
 {
 	struct request *prev = elv_former_request(q, rq);
 	if (prev)
 		return attempt_merge(q, prev, rq);
 	return 0;
 }
 /**
  * blk_attempt_remerge  - attempt to remerge active head with next request
  * @q:    The &request_queue_t belonging to the device
  * @rq:   The head request (usually)
  *
  * Description:
  *    For head-active devices, the queue can easily be unplugged so quickly
  *    that proper merging is not done on the front request. This may hurt
  *    performance greatly for some devices. The block layer cannot safely
  *    do merging on that first request for these queues, but the driver can
  *    call this function and make it happen any way. Only the driver knows
  *    when it is safe to do so.
  **/
 void blk_attempt_remerge(request_queue_t *q, struct request *rq)
 {
 	unsigned long flags;
 	spin_lock_irqsave(q->queue_lock, flags);
 	attempt_back_merge(q, rq);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 EXPORT_SYMBOL(blk_attempt_remerge);
 static int __make_request(request_queue_t *q, struct bio *bio)
 {
 	struct request *req;
 	int el_ret, rw, nr_sectors, cur_nr_sectors, barrier, err, sync;
 	unsigned short prio;
 	sector_t sector;
 	sector = bio->bi_sector;
 	nr_sectors = bio_sectors(bio);
 	cur_nr_sectors = bio_cur_sectors(bio);
 	prio = bio_prio(bio);
 	rw = bio_data_dir(bio);
 	sync = bio_sync(bio);
 	/*
 	 * low level driver can indicate that it wants pages above a
 	 * certain limit bounced to low memory (ie for highmem, or even
 	 * ISA dma in theory)
 	 */
 	blk_queue_bounce(q, &bio);
 	spin_lock_prefetch(q->queue_lock);
 	barrier = bio_barrier(bio);
 	if (unlikely(barrier) && (q->ordered == QUEUE_ORDERED_NONE)) {
 		err = -EOPNOTSUPP;
 		goto end_io;
 	}
 	spin_lock_irq(q->queue_lock);
 	if (unlikely(barrier) || elv_queue_empty(q))
 		goto get_rq;
 	el_ret = elv_merge(q, &req, bio);
 	switch (el_ret) {
 		case ELEVATOR_BACK_MERGE:
 			BUG_ON(!rq_mergeable(req));
 			if (!q->back_merge_fn(q, req, bio))
 				break;
 			req->biotail->bi_next = bio;
 			req->biotail = bio;
 			req->nr_sectors = req->hard_nr_sectors += nr_sectors;
 			req->ioprio = ioprio_best(req->ioprio, prio);
 			drive_stat_acct(req, nr_sectors, 0);
 			if (!attempt_back_merge(q, req))
 				elv_merged_request(q, req);
 			goto out;
 		case ELEVATOR_FRONT_MERGE:
 			BUG_ON(!rq_mergeable(req));
 			if (!q->front_merge_fn(q, req, bio))
 				break;
 			bio->bi_next = req->bio;
 			req->bio = bio;
 			/*
 			 * may not be valid. if the low level driver said
 			 * it didn't need a bounce buffer then it better
 			 * not touch req->buffer either...
 			 */
 			req->buffer = bio_data(bio);
 			req->current_nr_sectors = cur_nr_sectors;
 			req->hard_cur_sectors = cur_nr_sectors;
 			req->sector = req->hard_sector = sector;
 			req->nr_sectors = req->hard_nr_sectors += nr_sectors;
 			req->ioprio = ioprio_best(req->ioprio, prio);
 			drive_stat_acct(req, nr_sectors, 0);
 			if (!attempt_front_merge(q, req))
 				elv_merged_request(q, req);
 			goto out;
 		/* ELV_NO_MERGE: elevator says don't/can't merge. */
 		default:
 			;
 	}
 get_rq:
 	/*
 	 * Grab a free request. This is might sleep but can not fail.
 	 * Returns with the queue unlocked.
 	 */
 	req = get_request_wait(q, rw, bio);
 	/*
 	 * After dropping the lock and possibly sleeping here, our request
 	 * may now be mergeable after it had proven unmergeable (above).
 	 * We don't worry about that case for efficiency. It won't happen
 	 * often, and the elevators are able to handle it.
 	 */
 	req->flags |= REQ_CMD;
 	/*
 	 * inherit FAILFAST from bio (for read-ahead, and explicit FAILFAST)
 	 */
 	if (bio_rw_ahead(bio) || bio_failfast(bio))
 		req->flags |= REQ_FAILFAST;
 	/*
 	 * REQ_BARRIER implies no merging, but lets make it explicit
 	 */
 	if (unlikely(barrier))
 		req->flags |= (REQ_HARDBARRIER | REQ_NOMERGE);
 	req->errors = 0;
 	req->hard_sector = req->sector = sector;
 	req->hard_nr_sectors = req->nr_sectors = nr_sectors;
 	req->current_nr_sectors = req->hard_cur_sectors = cur_nr_sectors;
 	req->nr_phys_segments = bio_phys_segments(q, bio);
 	req->nr_hw_segments = bio_hw_segments(q, bio);
 	req->buffer = bio_data(bio);	/* see ->buffer comment above */
 	req->waiting = NULL;
 	req->bio = req->biotail = bio;
 	req->ioprio = prio;
 	req->rq_disk = bio->bi_bdev->bd_disk;
 	req->start_time = jiffies;
 	spin_lock_irq(q->queue_lock);
 	if (elv_queue_empty(q))
 		blk_plug_device(q);
 	add_request(q, req);
 out:
 	if (sync)
 		__generic_unplug_device(q);
 	spin_unlock_irq(q->queue_lock);
 	return 0;
 end_io:
 	bio_endio(bio, nr_sectors << 9, err);
 	return 0;
 }
 /*
  * If bio->bi_dev is a partition, remap the location
  */
 static inline void blk_partition_remap(struct bio *bio)
 {
 	struct block_device *bdev = bio->bi_bdev;
 	if (bdev != bdev->bd_contains) {
 		struct hd_struct *p = bdev->bd_part;
 		const int rw = bio_data_dir(bio);
 		p->sectors[rw] += bio_sectors(bio);
 		p->ios[rw]++;
 		bio->bi_sector += p->start_sect;
 		bio->bi_bdev = bdev->bd_contains;
 	}
 }
 static void handle_bad_sector(struct bio *bio)
 {
 	char b[BDEVNAME_SIZE];
 	printk(KERN_INFO "attempt to access beyond end of device\n");
 	printk(KERN_INFO "%s: rw=%ld, want=%Lu, limit=%Lu\n",
 			bdevname(bio->bi_bdev, b),
 			bio->bi_rw,
 			(unsigned long long)bio->bi_sector + bio_sectors(bio),
 			(long long)(bio->bi_bdev->bd_inode->i_size >> 9));
 	set_bit(BIO_EOF, &bio->bi_flags);
 }
 /**
  * generic_make_request: hand a buffer to its device driver for I/O
  * @bio:  The bio describing the location in memory and on the device.
  *
  * generic_make_request() is used to make I/O requests of block
  * devices. It is passed a &struct bio, which describes the I/O that needs
  * to be done.
  *
  * generic_make_request() does not return any status.  The
  * success/failure status of the request, along with notification of
  * completion, is delivered asynchronously through the bio->bi_end_io
  * function described (one day) else where.
  *
  * The caller of generic_make_request must make sure that bi_io_vec
  * are set to describe the memory buffer, and that bi_dev and bi_sector are
  * set to describe the device address, and the
  * bi_end_io and optionally bi_private are set to describe how
  * completion notification should be signaled.
  *
  * generic_make_request and the drivers it calls may use bi_next if this
  * bio happens to be merged with someone else, and may change bi_dev and
  * bi_sector for remaps as it sees fit.  So the values of these fields
  * should NOT be depended on after the call to generic_make_request.
  */
 void generic_make_request(struct bio *bio)
 {
 	request_queue_t *q;
 	sector_t maxsector;
 	int ret, nr_sectors = bio_sectors(bio);
 	might_sleep();
 	/* Test device or partition size, when known. */
 	maxsector = bio->bi_bdev->bd_inode->i_size >> 9;
 	if (maxsector) {
 		sector_t sector = bio->bi_sector;
 		if (maxsector < nr_sectors || maxsector - nr_sectors < sector) {
 			/*
 			 * This may well happen - the kernel calls bread()
 			 * without checking the size of the device, e.g., when
 			 * mounting a device.
 			 */
 			handle_bad_sector(bio);
 			goto end_io;
 		}
 	}
 	/*
 	 * Resolve the mapping until finished. (drivers are
 	 * still free to implement/resolve their own stacking
 	 * by explicitly returning 0)
 	 *
 	 * NOTE: we don't repeat the blk_size check for each new device.
 	 * Stacking drivers are expected to know what they are doing.
 	 */
 	do {
 		char b[BDEVNAME_SIZE];
 		q = bdev_get_queue(bio->bi_bdev);
 		if (!q) {
 			printk(KERN_ERR
 			       "generic_make_request: Trying to access "
 				"nonexistent block-device %s (%Lu)\n",
 				bdevname(bio->bi_bdev, b),
 				(long long) bio->bi_sector);
 end_io:
 			bio_endio(bio, bio->bi_size, -EIO);
 			break;
 		}
 		if (unlikely(bio_sectors(bio) > q->max_hw_sectors)) {
 			printk("bio too big device %s (%u > %u)\n",
 				bdevname(bio->bi_bdev, b),
 				bio_sectors(bio),
 				q->max_hw_sectors);
 			goto end_io;
 		}
 		if (unlikely(test_bit(QUEUE_FLAG_DEAD, &q->queue_flags)))
 			goto end_io;
 		/*
 		 * If this device has partitions, remap block n
 		 * of partition p to block n+start(p) of the disk.
 		 */
 		blk_partition_remap(bio);
 		ret = q->make_request_fn(q, bio);
 	} while (ret);
 }
 EXPORT_SYMBOL(generic_make_request);
 /**
  * submit_bio: submit a bio to the block device layer for I/O
  * @rw: whether to %READ or %WRITE, or maybe to %READA (read ahead)
  * @bio: The &struct bio which describes the I/O
  *
  * submit_bio() is very similar in purpose to generic_make_request(), and
  * uses that function to do most of the work. Both are fairly rough
  * interfaces, @bio must be presetup and ready for I/O.
  *
  */
 void submit_bio(int rw, struct bio *bio)
 {
 	int count = bio_sectors(bio);
 	BIO_BUG_ON(!bio->bi_size);
 	BIO_BUG_ON(!bio->bi_io_vec);
 	bio->bi_rw |= rw;
 	if (rw & WRITE)
 		mod_page_state(pgpgout, count);
 	else
 		mod_page_state(pgpgin, count);
 	if (unlikely(block_dump)) {
 		char b[BDEVNAME_SIZE];
 		printk(KERN_DEBUG "%s(%d): %s block %Lu on %s\n",
 			current->comm, current->pid,
 			(rw & WRITE) ? "WRITE" : "READ",
 			(unsigned long long)bio->bi_sector,
 			bdevname(bio->bi_bdev,b));
 	}
 	generic_make_request(bio);
 }
 EXPORT_SYMBOL(submit_bio);
 static void blk_recalc_rq_segments(struct request *rq)
 {
 	struct bio *bio, *prevbio = NULL;
 	int nr_phys_segs, nr_hw_segs;
 	unsigned int phys_size, hw_size;
 	request_queue_t *q = rq->q;
 	if (!rq->bio)
 		return;
 	phys_size = hw_size = nr_phys_segs = nr_hw_segs = 0;
 	rq_for_each_bio(bio, rq) {
 		/* Force bio hw/phys segs to be recalculated. */
 		bio->bi_flags &= ~(1 << BIO_SEG_VALID);
 		nr_phys_segs += bio_phys_segments(q, bio);
 		nr_hw_segs += bio_hw_segments(q, bio);
 		if (prevbio) {
 			int pseg = phys_size + prevbio->bi_size + bio->bi_size;
 			int hseg = hw_size + prevbio->bi_size + bio->bi_size;
 			if (blk_phys_contig_segment(q, prevbio, bio) &&
 			    pseg <= q->max_segment_size) {
 				nr_phys_segs--;
 				phys_size += prevbio->bi_size + bio->bi_size;
 			} else
 				phys_size = 0;
 			if (blk_hw_contig_segment(q, prevbio, bio) &&
 			    hseg <= q->max_segment_size) {
 				nr_hw_segs--;
 				hw_size += prevbio->bi_size + bio->bi_size;
 			} else
 				hw_size = 0;
 		}
 		prevbio = bio;
 	}
 	rq->nr_phys_segments = nr_phys_segs;
 	rq->nr_hw_segments = nr_hw_segs;
 }
 static void blk_recalc_rq_sectors(struct request *rq, int nsect)
 {
 	if (blk_fs_request(rq)) {
 		rq->hard_sector += nsect;
 		rq->hard_nr_sectors -= nsect;
 		/*
 		 * Move the I/O submission pointers ahead if required.
 		 */
 		if ((rq->nr_sectors >= rq->hard_nr_sectors) &&
 		    (rq->sector <= rq->hard_sector)) {
 			rq->sector = rq->hard_sector;
 			rq->nr_sectors = rq->hard_nr_sectors;
 			rq->hard_cur_sectors = bio_cur_sectors(rq->bio);
 			rq->current_nr_sectors = rq->hard_cur_sectors;
 			rq->buffer = bio_data(rq->bio);
 		}
 		/*
 		 * if total number of sectors is less than the first segment
 		 * size, something has gone terribly wrong
 		 */
 		if (rq->nr_sectors < rq->current_nr_sectors) {
 			printk("blk: request botched\n");
 			rq->nr_sectors = rq->current_nr_sectors;
 		}
 	}
 }
 static int __end_that_request_first(struct request *req, int uptodate,
 				    int nr_bytes)
 {
 	int total_bytes, bio_nbytes, error, next_idx = 0;
 	struct bio *bio;
 	/*
 	 * extend uptodate bool to allow < 0 value to be direct io error
 	 */
 	error = 0;
 	if (end_io_error(uptodate))
 		error = !uptodate ? -EIO : uptodate;
 	/*
 	 * for a REQ_BLOCK_PC request, we want to carry any eventual
 	 * sense key with us all the way through
 	 */
 	if (!blk_pc_request(req))
 		req->errors = 0;
 	if (!uptodate) {
 		if (blk_fs_request(req) && !(req->flags & REQ_QUIET))
 			printk("end_request: I/O error, dev %s, sector %llu\n",
 				req->rq_disk ? req->rq_disk->disk_name : "?",
 				(unsigned long long)req->sector);
 	}
 	if (blk_fs_request(req) && req->rq_disk) {
 		const int rw = rq_data_dir(req);
 		__disk_stat_add(req->rq_disk, sectors[rw], nr_bytes >> 9);
 	}
 	total_bytes = bio_nbytes = 0;
 	while ((bio = req->bio) != NULL) {
 		int nbytes;
 		if (nr_bytes >= bio->bi_size) {
 			req->bio = bio->bi_next;
 			nbytes = bio->bi_size;
 			bio_endio(bio, nbytes, error);
 			next_idx = 0;
 			bio_nbytes = 0;
 		} else {
 			int idx = bio->bi_idx + next_idx;
 			if (unlikely(bio->bi_idx >= bio->bi_vcnt)) {
 				blk_dump_rq_flags(req, "__end_that");
 				printk("%s: bio idx %d >= vcnt %d\n",
 						__FUNCTION__,
 						bio->bi_idx, bio->bi_vcnt);
 				break;
 			}
 			nbytes = bio_iovec_idx(bio, idx)->bv_len;
 			BIO_BUG_ON(nbytes > bio->bi_size);
 			/*
 			 * not a complete bvec done
 			 */
 			if (unlikely(nbytes > nr_bytes)) {
 				bio_nbytes += nr_bytes;
 				total_bytes += nr_bytes;
 				break;
 			}
 			/*
 			 * advance to the next vector
 			 */
 			next_idx++;
 			bio_nbytes += nbytes;
 		}
 		total_bytes += nbytes;
 		nr_bytes -= nbytes;
 		if ((bio = req->bio)) {
 			/*
 			 * end more in this run, or just return 'not-done'
 			 */
 			if (unlikely(nr_bytes <= 0))
 				break;
 		}
 	}
 	/*
 	 * completely done
 	 */
 	if (!req->bio)
 		return 0;
 	/*
 	 * if the request wasn't completed, update state
 	 */
 	if (bio_nbytes) {
 		bio_endio(bio, bio_nbytes, error);
 		bio->bi_idx += next_idx;
 		bio_iovec(bio)->bv_offset += nr_bytes;
 		bio_iovec(bio)->bv_len -= nr_bytes;
 	}
 	blk_recalc_rq_sectors(req, total_bytes >> 9);
 	blk_recalc_rq_segments(req);
 	return 1;
 }
 /**
  * end_that_request_first - end I/O on a request
  * @req:      the request being processed
  * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error
  * @nr_sectors: number of sectors to end I/O on
  *
  * Description:
  *     Ends I/O on a number of sectors attached to @req, and sets it up
  *     for the next range of segments (if any) in the cluster.
  *
  * Return:
  *     0 - we are done with this request, call end_that_request_last()
  *     1 - still buffers pending for this request
  **/
 int end_that_request_first(struct request *req, int uptodate, int nr_sectors)
 {
 	return __end_that_request_first(req, uptodate, nr_sectors << 9);
 }
 EXPORT_SYMBOL(end_that_request_first);
 /**
  * end_that_request_chunk - end I/O on a request
  * @req:      the request being processed
  * @uptodate: 1 for success, 0 for I/O error, < 0 for specific error
  * @nr_bytes: number of bytes to complete
  *
  * Description:
  *     Ends I/O on a number of bytes attached to @req, and sets it up
  *     for the next range of segments (if any). Like end_that_request_first(),
  *     but deals with bytes instead of sectors.
  *
  * Return:
  *     0 - we are done with this request, call end_that_request_last()
  *     1 - still buffers pending for this request
  **/
 int end_that_request_chunk(struct request *req, int uptodate, int nr_bytes)
 {
 	return __end_that_request_first(req, uptodate, nr_bytes);
 }
 EXPORT_SYMBOL(end_that_request_chunk);
 /*
  * queue lock must be held
  */
 void end_that_request_last(struct request *req)
 {
 	struct gendisk *disk = req->rq_disk;
 	if (unlikely(laptop_mode) && blk_fs_request(req))
 		laptop_io_completion();
 	if (disk && blk_fs_request(req)) {
 		unsigned long duration = jiffies - req->start_time;
 		const int rw = rq_data_dir(req);
 		__disk_stat_inc(disk, ios[rw]);
 		__disk_stat_add(disk, ticks[rw], duration);
 		disk_round_stats(disk);
 		disk->in_flight--;
 	}
 	if (req->end_io)
 		req->end_io(req);
 	else
 		__blk_put_request(req->q, req);
 }
 EXPORT_SYMBOL(end_that_request_last);
 void end_request(struct request *req, int uptodate)
 {
 	if (!end_that_request_first(req, uptodate, req->hard_cur_sectors)) {
 		add_disk_randomness(req->rq_disk);
 		blkdev_dequeue_request(req);
 		end_that_request_last(req);
 	}
 }
 EXPORT_SYMBOL(end_request);
 void blk_rq_bio_prep(request_queue_t *q, struct request *rq, struct bio *bio)
 {
 	/* first three bits are identical in rq->flags and bio->bi_rw */
 	rq->flags |= (bio->bi_rw & 7);
 	rq->nr_phys_segments = bio_phys_segments(q, bio);
 	rq->nr_hw_segments = bio_hw_segments(q, bio);
 	rq->current_nr_sectors = bio_cur_sectors(bio);
 	rq->hard_cur_sectors = rq->current_nr_sectors;
 	rq->hard_nr_sectors = rq->nr_sectors = bio_sectors(bio);
 	rq->buffer = bio_data(bio);
 	rq->bio = rq->biotail = bio;
 }
 EXPORT_SYMBOL(blk_rq_bio_prep);
 int kblockd_schedule_work(struct work_struct *work)
 {
 	return queue_work(kblockd_workqueue, work);
 }
 EXPORT_SYMBOL(kblockd_schedule_work);
 void kblockd_flush(void)
 {
 	flush_workqueue(kblockd_workqueue);
 }
 EXPORT_SYMBOL(kblockd_flush);
 int __init blk_dev_init(void)
 {
 	kblockd_workqueue = create_workqueue("kblockd");
 	if (!kblockd_workqueue)
 		panic("Failed to create kblockd\n");
 	request_cachep = kmem_cache_create("blkdev_requests",
 			sizeof(struct request), 0, SLAB_PANIC, NULL, NULL);
 	requestq_cachep = kmem_cache_create("blkdev_queue",
 			sizeof(request_queue_t), 0, SLAB_PANIC, NULL, NULL);
 	iocontext_cachep = kmem_cache_create("blkdev_ioc",
 			sizeof(struct io_context), 0, SLAB_PANIC, NULL, NULL);
 	blk_max_low_pfn = max_low_pfn;
 	blk_max_pfn = max_pfn;
 	return 0;
 }
 /*
  * IO Context helper functions
  */
 void put_io_context(struct io_context *ioc)
 {
 	if (ioc == NULL)
 		return;
 	BUG_ON(atomic_read(&ioc->refcount) == 0);
 	if (atomic_dec_and_test(&ioc->refcount)) {
 		if (ioc->aic && ioc->aic->dtor)
 			ioc->aic->dtor(ioc->aic);
 		if (ioc->cic && ioc->cic->dtor)
 			ioc->cic->dtor(ioc->cic);
 		kmem_cache_free(iocontext_cachep, ioc);
 	}
 }
 EXPORT_SYMBOL(put_io_context);
 /* Called by the exitting task */
 void exit_io_context(void)
 {
 	unsigned long flags;
 	struct io_context *ioc;
 	local_irq_save(flags);
 	task_lock(current);
 	ioc = current->io_context;
 	current->io_context = NULL;
 	ioc->task = NULL;
 	task_unlock(current);
 	local_irq_restore(flags);
 	if (ioc->aic && ioc->aic->exit)
 		ioc->aic->exit(ioc->aic);
 	if (ioc->cic && ioc->cic->exit)
 		ioc->cic->exit(ioc->cic);
 	put_io_context(ioc);
 }
 /*
  * If the current task has no IO context then create one and initialise it.
  * Otherwise, return its existing IO context.
  *
  * This returned IO context doesn't have a specifically elevated refcount,
  * but since the current task itself holds a reference, the context can be
  * used in general code, so long as it stays within `current` context.
  */
 struct io_context *current_io_context(gfp_t gfp_flags)
 {
 	struct task_struct *tsk = current;
 	struct io_context *ret;
 	ret = tsk->io_context;
 	if (likely(ret))
 		return ret;
 	ret = kmem_cache_alloc(iocontext_cachep, gfp_flags);
 	if (ret) {
 		atomic_set(&ret->refcount, 1);
 		ret->task = current;
 		ret->set_ioprio = NULL;
 		ret->last_waited = jiffies; /* doesn't matter... */
 		ret->nr_batch_requests = 0; /* because this is 0 */
 		ret->aic = NULL;
 		ret->cic = NULL;
 		tsk->io_context = ret;
 	}
 	return ret;
 }
 EXPORT_SYMBOL(current_io_context);
 /*
  * If the current task has no IO context then create one and initialise it.
  * If it does have a context, take a ref on it.
  *
  * This is always called in the context of the task which submitted the I/O.
  */
 struct io_context *get_io_context(gfp_t gfp_flags)
 {
 	struct io_context *ret;
 	ret = current_io_context(gfp_flags);
 	if (likely(ret))
 		atomic_inc(&ret->refcount);
 	return ret;
 }
 EXPORT_SYMBOL(get_io_context);
 void copy_io_context(struct io_context **pdst, struct io_context **psrc)
 {
 	struct io_context *src = *psrc;
 	struct io_context *dst = *pdst;
 	if (src) {
 		BUG_ON(atomic_read(&src->refcount) == 0);
 		atomic_inc(&src->refcount);
 		put_io_context(dst);
 		*pdst = src;
 	}
 }
 EXPORT_SYMBOL(copy_io_context);
 void swap_io_context(struct io_context **ioc1, struct io_context **ioc2)
 {
 	struct io_context *temp;
 	temp = *ioc1;
 	*ioc1 = *ioc2;
 	*ioc2 = temp;
 }
 EXPORT_SYMBOL(swap_io_context);
 /*
  * sysfs parts below
  */
 struct queue_sysfs_entry {
 	struct attribute attr;
 	ssize_t (*show)(struct request_queue *, char *);
 	ssize_t (*store)(struct request_queue *, const char *, size_t);
 };
 static ssize_t
 queue_var_show(unsigned int var, char *page)
 {
 	return sprintf(page, "%d\n", var);
 }
 static ssize_t
 queue_var_store(unsigned long *var, const char *page, size_t count)
 {
 	char *p = (char *) page;
 	*var = simple_strtoul(p, &p, 10);
 	return count;
 }
 static ssize_t queue_requests_show(struct request_queue *q, char *page)
 {
 	return queue_var_show(q->nr_requests, (page));
 }
 static ssize_t
 queue_requests_store(struct request_queue *q, const char *page, size_t count)
 {
 	struct request_list *rl = &q->rq;
 	int ret = queue_var_store(&q->nr_requests, page, count);
 	if (q->nr_requests < BLKDEV_MIN_RQ)
 		q->nr_requests = BLKDEV_MIN_RQ;
 	blk_queue_congestion_threshold(q);
 	if (rl->count[READ] >= queue_congestion_on_threshold(q))
 		set_queue_congested(q, READ);
 	else if (rl->count[READ] < queue_congestion_off_threshold(q))
 		clear_queue_congested(q, READ);
 	if (rl->count[WRITE] >= queue_congestion_on_threshold(q))
 		set_queue_congested(q, WRITE);
 	else if (rl->count[WRITE] < queue_congestion_off_threshold(q))
 		clear_queue_congested(q, WRITE);
 	if (rl->count[READ] >= q->nr_requests) {
 		blk_set_queue_full(q, READ);
 	} else if (rl->count[READ]+1 <= q->nr_requests) {
 		blk_clear_queue_full(q, READ);
 		wake_up(&rl->wait[READ]);
 	}
 	if (rl->count[WRITE] >= q->nr_requests) {
 		blk_set_queue_full(q, WRITE);
 	} else if (rl->count[WRITE]+1 <= q->nr_requests) {
 		blk_clear_queue_full(q, WRITE);
 		wake_up(&rl->wait[WRITE]);
 	}
 	return ret;
 }
 static ssize_t queue_ra_show(struct request_queue *q, char *page)
 {
 	int ra_kb = q->backing_dev_info.ra_pages << (PAGE_CACHE_SHIFT - 10);
 	return queue_var_show(ra_kb, (page));
 }
 static ssize_t
 queue_ra_store(struct request_queue *q, const char *page, size_t count)
 {
 	unsigned long ra_kb;
 	ssize_t ret = queue_var_store(&ra_kb, page, count);
 	spin_lock_irq(q->queue_lock);
 	if (ra_kb > (q->max_sectors >> 1))
 		ra_kb = (q->max_sectors >> 1);
 	q->backing_dev_info.ra_pages = ra_kb >> (PAGE_CACHE_SHIFT - 10);
 	spin_unlock_irq(q->queue_lock);
 	return ret;
 }
 static ssize_t queue_max_sectors_show(struct request_queue *q, char *page)
 {
 	int max_sectors_kb = q->max_sectors >> 1;
 	return queue_var_show(max_sectors_kb, (page));
 }
 static ssize_t
 queue_max_sectors_store(struct request_queue *q, const char *page, size_t count)
 {
 	unsigned long max_sectors_kb,
 			max_hw_sectors_kb = q->max_hw_sectors >> 1,
 			page_kb = 1 << (PAGE_CACHE_SHIFT - 10);
 	ssize_t ret = queue_var_store(&max_sectors_kb, page, count);
 	int ra_kb;
 	if (max_sectors_kb > max_hw_sectors_kb || max_sectors_kb < page_kb)
 		return -EINVAL;
 	/*
 	 * Take the queue lock to update the readahead and max_sectors
 	 * values synchronously:
 	 */
 	spin_lock_irq(q->queue_lock);
 	/*
 	 * Trim readahead window as well, if necessary:
 	 */
 	ra_kb = q->backing_dev_info.ra_pages << (PAGE_CACHE_SHIFT - 10);
 	if (ra_kb > max_sectors_kb)
 		q->backing_dev_info.ra_pages =
 				max_sectors_kb >> (PAGE_CACHE_SHIFT - 10);
 	q->max_sectors = max_sectors_kb << 1;
 	spin_unlock_irq(q->queue_lock);
 	return ret;
 }
 static ssize_t queue_max_hw_sectors_show(struct request_queue *q, char *page)
 {
 	int max_hw_sectors_kb = q->max_hw_sectors >> 1;
 	return queue_var_show(max_hw_sectors_kb, (page));
 }
 static struct queue_sysfs_entry queue_requests_entry = {
 	.attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR },
 	.show = queue_requests_show,
 	.store = queue_requests_store,
 };
 static struct queue_sysfs_entry queue_ra_entry = {
 	.attr = {.name = "read_ahead_kb", .mode = S_IRUGO | S_IWUSR },
 	.show = queue_ra_show,
 	.store = queue_ra_store,
 };
 static struct queue_sysfs_entry queue_max_sectors_entry = {
 	.attr = {.name = "max_sectors_kb", .mode = S_IRUGO | S_IWUSR },
 	.show = queue_max_sectors_show,
 	.store = queue_max_sectors_store,
 };
 static struct queue_sysfs_entry queue_max_hw_sectors_entry = {
 	.attr = {.name = "max_hw_sectors_kb", .mode = S_IRUGO },
 	.show = queue_max_hw_sectors_show,
 };
 static struct queue_sysfs_entry queue_iosched_entry = {
 	.attr = {.name = "scheduler", .mode = S_IRUGO | S_IWUSR },
 	.show = elv_iosched_show,
 	.store = elv_iosched_store,
 };
 static struct attribute *default_attrs[] = {
 	&queue_requests_entry.attr,
 	&queue_ra_entry.attr,
 	&queue_max_hw_sectors_entry.attr,
 	&queue_max_sectors_entry.attr,
 	&queue_iosched_entry.attr,
 	NULL,
 };
 #define to_queue(atr) container_of((atr), struct queue_sysfs_entry, attr)
 static ssize_t
 queue_attr_show(struct kobject *kobj, struct attribute *attr, char *page)
 {
 	struct queue_sysfs_entry *entry = to_queue(attr);
 	struct request_queue *q;
 	q = container_of(kobj, struct request_queue, kobj);
 	if (!entry->show)
 		return -EIO;
 	return entry->show(q, page);
 }
 static ssize_t
 queue_attr_store(struct kobject *kobj, struct attribute *attr,
 		    const char *page, size_t length)
 {
 	struct queue_sysfs_entry *entry = to_queue(attr);
 	struct request_queue *q;
 	q = container_of(kobj, struct request_queue, kobj);
 	if (!entry->store)
 		return -EIO;
 	return entry->store(q, page, length);
 }
 static struct sysfs_ops queue_sysfs_ops = {
 	.show	= queue_attr_show,
 	.store	= queue_attr_store,
 };
 static struct kobj_type queue_ktype = {
 	.sysfs_ops	= &queue_sysfs_ops,
 	.default_attrs	= default_attrs,
 };
 int blk_register_queue(struct gendisk *disk)
 {
 	int ret;
 	request_queue_t *q = disk->queue;
 	if (!q || !q->request_fn)
 		return -ENXIO;
 	q->kobj.parent = kobject_get(&disk->kobj);
 	if (!q->kobj.parent)
 		return -EBUSY;
 	snprintf(q->kobj.name, KOBJ_NAME_LEN, "%s", "queue");
 	q->kobj.ktype = &queue_ktype;
 	ret = kobject_register(&q->kobj);
 	if (ret < 0)
 		return ret;
 	ret = elv_register_queue(q);
 	if (ret) {
 		kobject_unregister(&q->kobj);
 		return ret;
 	}
 	return 0;
 }
 void blk_unregister_queue(struct gendisk *disk)
 {
 	request_queue_t *q = disk->queue;
 	if (q && q->request_fn) {
 		elv_unregister_queue(q);
 		kobject_unregister(&q->kobj);
 		kobject_put(&disk->kobj);
 	}
 }

fs/fs-writeback.c

Diff comments View file @ b8887e6

 /*
  * fs/fs-writeback.c
  *
  * Copyright (C) 2002, Linus Torvalds.
  *
  * Contains all the functions related to writing back and waiting
  * upon dirty inodes against superblocks, and writing back dirty
  * pages against inodes.  ie: data writeback.  Writeout of the
  * inode itself is not handled here.
  *
  * 10Apr2002	akpm@zip.com.au
  *		Split out of fs/inode.c
  *		Additions for address_space-based writeback
  */
 #include <linux/kernel.h>
 #include <linux/spinlock.h>
 #include <linux/sched.h>
 #include <linux/fs.h>
 #include <linux/mm.h>
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
 #include <linux/buffer_head.h>
 extern struct super_block *blockdev_superblock;
 /**
  *	__mark_inode_dirty -	internal function
  *	@inode: inode to mark
  *	@flags: what kind of dirty (i.e. I_DIRTY_SYNC)
  *	Mark an inode as dirty. Callers should use mark_inode_dirty or
  *  	mark_inode_dirty_sync.
  *
  * Put the inode on the super block's dirty list.
  *
  * CAREFUL! We mark it dirty unconditionally, but move it onto the
  * dirty list only if it is hashed or if it refers to a blockdev.
  * If it was not hashed, it will never be added to the dirty list
  * even if it is later hashed, as it will have been marked dirty already.
  *
  * In short, make sure you hash any inodes _before_ you start marking
  * them dirty.
  *
  * This function *must* be atomic for the I_DIRTY_PAGES case -
  * set_page_dirty() is called under spinlock in several places.
  *
  * Note that for blockdevs, inode->dirtied_when represents the dirtying time of
  * the block-special inode (/dev/hda1) itself.  And the ->dirtied_when field of
  * the kernel-internal blockdev inode represents the dirtying time of the
  * blockdev's pages.  This is why for I_DIRTY_PAGES we always use
  * page->mapping->host, so the page-dirtying time is recorded in the internal
  * blockdev inode.
  */
 void __mark_inode_dirty(struct inode *inode, int flags)
 {
 	struct super_block *sb = inode->i_sb;
 	/*
 	 * Don't do this for I_DIRTY_PAGES - that doesn't actually
 	 * dirty the inode itself
 	 */
 	if (flags & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) {
 		if (sb->s_op->dirty_inode)
 			sb->s_op->dirty_inode(inode);
 	}
 	/*
 	 * make sure that changes are seen by all cpus before we test i_state
 	 * -- mikulas
 	 */
 	smp_mb();
 	/* avoid the locking if we can */
 	if ((inode->i_state & flags) == flags)
 		return;
 	if (unlikely(block_dump)) {
 		struct dentry *dentry = NULL;
 		const char *name = "?";
 		if (!list_empty(&inode->i_dentry)) {
 			dentry = list_entry(inode->i_dentry.next,
 					    struct dentry, d_alias);
 			if (dentry && dentry->d_name.name)
 				name = (const char *) dentry->d_name.name;
 		}
 		if (inode->i_ino || strcmp(inode->i_sb->s_id, "bdev"))
 			printk(KERN_DEBUG
 			       "%s(%d): dirtied inode %lu (%s) on %s\n",
 			       current->comm, current->pid, inode->i_ino,
 			       name, inode->i_sb->s_id);
 	}
 	spin_lock(&inode_lock);
 	if ((inode->i_state & flags) != flags) {
 		const int was_dirty = inode->i_state & I_DIRTY;
 		inode->i_state |= flags;
 		/*
 		 * If the inode is locked, just update its dirty state.
 		 * The unlocker will place the inode on the appropriate
 		 * superblock list, based upon its state.
 		 */
 		if (inode->i_state & I_LOCK)
 			goto out;
 		/*
 		 * Only add valid (hashed) inodes to the superblock's
 		 * dirty list.  Add blockdev inodes as well.
 		 */
 		if (!S_ISBLK(inode->i_mode)) {
 			if (hlist_unhashed(&inode->i_hash))
 				goto out;
 		}
 		if (inode->i_state & (I_FREEING|I_CLEAR))
 			goto out;
 		/*
 		 * If the inode was already on s_dirty or s_io, don't
 		 * reposition it (that would break s_dirty time-ordering).
 		 */
 		if (!was_dirty) {
 			inode->dirtied_when = jiffies;
 			list_move(&inode->i_list, &sb->s_dirty);
 		}
 	}
 out:
 	spin_unlock(&inode_lock);
 }
 EXPORT_SYMBOL(__mark_inode_dirty);
 static int write_inode(struct inode *inode, int sync)
 {
 	if (inode->i_sb->s_op->write_inode && !is_bad_inode(inode))
 		return inode->i_sb->s_op->write_inode(inode, sync);
 	return 0;
 }
 /*
  * Write a single inode's dirty pages and inode data out to disk.
  * If `wait' is set, wait on the writeout.
  *
  * The whole writeout design is quite complex and fragile.  We want to avoid
  * starvation of particular inodes when others are being redirtied, prevent
  * livelocks, etc.
  *
  * Called under inode_lock.
  */
 static int
 __sync_single_inode(struct inode *inode, struct writeback_control *wbc)
 {
 	unsigned dirty;
 	struct address_space *mapping = inode->i_mapping;
 	struct super_block *sb = inode->i_sb;
 	int wait = wbc->sync_mode == WB_SYNC_ALL;
 	int ret;
 	BUG_ON(inode->i_state & I_LOCK);
 	/* Set I_LOCK, reset I_DIRTY */
 	dirty = inode->i_state & I_DIRTY;
 	inode->i_state |= I_LOCK;
 	inode->i_state &= ~I_DIRTY;
 	spin_unlock(&inode_lock);
 	ret = do_writepages(mapping, wbc);
 	/* Don't write the inode if only I_DIRTY_PAGES was set */
 	if (dirty & (I_DIRTY_SYNC | I_DIRTY_DATASYNC)) {
 		int err = write_inode(inode, wait);
 		if (ret == 0)
 			ret = err;
 	}
 	if (wait) {
 		int err = filemap_fdatawait(mapping);
 		if (ret == 0)
 			ret = err;
 	}
 	spin_lock(&inode_lock);
 	inode->i_state &= ~I_LOCK;
 	if (!(inode->i_state & I_FREEING)) {
 		if (!(inode->i_state & I_DIRTY) &&
 		    mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
 			/*
 			 * We didn't write back all the pages.  nfs_writepages()
 			 * sometimes bales out without doing anything. Redirty
 			 * the inode.  It is still on sb->s_io.
 			 */
 			if (wbc->for_kupdate) {
 				/*
 				 * For the kupdate function we leave the inode
 				 * at the head of sb_dirty so it will get more
 				 * writeout as soon as the queue becomes
 				 * uncongested.
 				 */
 				inode->i_state |= I_DIRTY_PAGES;
 				list_move_tail(&inode->i_list, &sb->s_dirty);
 			} else {
 				/*
 				 * Otherwise fully redirty the inode so that
 				 * other inodes on this superblock will get some
 				 * writeout.  Otherwise heavy writing to one
 				 * file would indefinitely suspend writeout of
 				 * all the other files.
 				 */
 				inode->i_state |= I_DIRTY_PAGES;
 				inode->dirtied_when = jiffies;
 				list_move(&inode->i_list, &sb->s_dirty);
 			}
 		} else if (inode->i_state & I_DIRTY) {
 			/*
 			 * Someone redirtied the inode while were writing back
 			 * the pages.
 			 */
 			list_move(&inode->i_list, &sb->s_dirty);
 		} else if (atomic_read(&inode->i_count)) {
 			/*
 			 * The inode is clean, inuse
 			 */
 			list_move(&inode->i_list, &inode_in_use);
 		} else {
 			/*
 			 * The inode is clean, unused
 			 */
 			list_move(&inode->i_list, &inode_unused);
 		}
 	}
 	wake_up_inode(inode);
 	return ret;
 }
 /*
  * Write out an inode's dirty pages.  Called under inode_lock.  Either the
  * caller has ref on the inode (either via __iget or via syscall against an fd)
  * or the inode has I_WILL_FREE set (via generic_forget_inode)
  */
 static int
 __writeback_single_inode(struct inode *inode, struct writeback_control *wbc)
 {
 	wait_queue_head_t *wqh;
 	if (!atomic_read(&inode->i_count))
 		WARN_ON(!(inode->i_state & (I_WILL_FREE|I_FREEING)));
 	else
 		WARN_ON(inode->i_state & I_WILL_FREE);
 	if ((wbc->sync_mode != WB_SYNC_ALL) && (inode->i_state & I_LOCK)) {
 		list_move(&inode->i_list, &inode->i_sb->s_dirty);
 		return 0;
 	}
 	/*
 	 * It's a data-integrity sync.  We must wait.
 	 */
 	if (inode->i_state & I_LOCK) {
 		DEFINE_WAIT_BIT(wq, &inode->i_state, __I_LOCK);
 		wqh = bit_waitqueue(&inode->i_state, __I_LOCK);
 		do {
 			spin_unlock(&inode_lock);
 			__wait_on_bit(wqh, &wq, inode_wait,
 							TASK_UNINTERRUPTIBLE);
 			spin_lock(&inode_lock);
 		} while (inode->i_state & I_LOCK);
 	}
 	return __sync_single_inode(inode, wbc);
 }
 /*
  * Write out a superblock's list of dirty inodes.  A wait will be performed
  * upon no inodes, all inodes or the final one, depending upon sync_mode.
  *
  * If older_than_this is non-NULL, then only write out inodes which
  * had their first dirtying at a time earlier than *older_than_this.
  *
  * If we're a pdlfush thread, then implement pdflush collision avoidance
  * against the entire list.
  *
  * WB_SYNC_HOLD is a hack for sys_sync(): reattach the inode to sb->s_dirty so
  * that it can be located for waiting on in __writeback_single_inode().
  *
  * Called under inode_lock.
  *
  * If `bdi' is non-zero then we're being asked to writeback a specific queue.
  * This function assumes that the blockdev superblock's inodes are backed by
  * a variety of queues, so all inodes are searched.  For other superblocks,
  * assume that all inodes are backed by the same queue.
  *
  * FIXME: this linear search could get expensive with many fileystems.  But
  * how to fix?  We need to go from an address_space to all inodes which share
  * a queue with that address_space.  (Easy: have a global "dirty superblocks"
  * list).
  *
  * The inodes to be written are parked on sb->s_io.  They are moved back onto
  * sb->s_dirty as they are selected for writing.  This way, none can be missed
  * on the writer throttling path, and we get decent balancing between many
  * throttled threads: we don't want them all piling up on __wait_on_inode.
  */
 static void
 sync_sb_inodes(struct super_block *sb, struct writeback_control *wbc)
 {
 	const unsigned long start = jiffies;	/* livelock avoidance */
 	if (!wbc->for_kupdate || list_empty(&sb->s_io))
 		list_splice_init(&sb->s_dirty, &sb->s_io);
 	while (!list_empty(&sb->s_io)) {
 		struct inode *inode = list_entry(sb->s_io.prev,
 						struct inode, i_list);
 		struct address_space *mapping = inode->i_mapping;
 		struct backing_dev_info *bdi = mapping->backing_dev_info;
 		long pages_skipped;
 		if (!bdi_cap_writeback_dirty(bdi)) {
 			list_move(&inode->i_list, &sb->s_dirty);
 			if (sb == blockdev_superblock) {
 				/*
 				 * Dirty memory-backed blockdev: the ramdisk
 				 * driver does this.  Skip just this inode
 				 */
 				continue;
 			}
 			/*
 			 * Dirty memory-backed inode against a filesystem other
 			 * than the kernel-internal bdev filesystem.  Skip the
 			 * entire superblock.
 			 */
 			break;
 		}
 		if (wbc->nonblocking && bdi_write_congested(bdi)) {
 			wbc->encountered_congestion = 1;
 			if (sb != blockdev_superblock)
 				break;		/* Skip a congested fs */
 			list_move(&inode->i_list, &sb->s_dirty);
 			continue;		/* Skip a congested blockdev */
 		}
 		if (wbc->bdi && bdi != wbc->bdi) {
 			if (sb != blockdev_superblock)
 				break;		/* fs has the wrong queue */
 			list_move(&inode->i_list, &sb->s_dirty);
 			continue;		/* blockdev has wrong queue */
 		}
 		/* Was this inode dirtied after sync_sb_inodes was called? */
 		if (time_after(inode->dirtied_when, start))
 			break;
 		/* Was this inode dirtied too recently? */
 		if (wbc->older_than_this && time_after(inode->dirtied_when,
 						*wbc->older_than_this))
 			break;
 		/* Is another pdflush already flushing this queue? */
 		if (current_is_pdflush() && !writeback_acquire(bdi))
 			break;
 		BUG_ON(inode->i_state & I_FREEING);
 		__iget(inode);
 		pages_skipped = wbc->pages_skipped;
 		__writeback_single_inode(inode, wbc);
 		if (wbc->sync_mode == WB_SYNC_HOLD) {
 			inode->dirtied_when = jiffies;
 			list_move(&inode->i_list, &sb->s_dirty);
 		}
 		if (current_is_pdflush())
 			writeback_release(bdi);
 		if (wbc->pages_skipped != pages_skipped) {
 			/*
 			 * writeback is not making progress due to locked
 			 * buffers.  Skip this inode for now.
 			 */
 			list_move(&inode->i_list, &sb->s_dirty);
 		}
 		spin_unlock(&inode_lock);
 		cond_resched();
 		iput(inode);
 		spin_lock(&inode_lock);
 		if (wbc->nr_to_write <= 0)
 			break;
 	}
 	return;		/* Leave any unwritten inodes on s_io */
 }
 /*
  * Start writeback of dirty pagecache data against all unlocked inodes.
  *
  * Note:
  * We don't need to grab a reference to superblock here. If it has non-empty
  * ->s_dirty it's hadn't been killed yet and kill_super() won't proceed
  * past sync_inodes_sb() until both the ->s_dirty and ->s_io lists are
  * empty. Since __sync_single_inode() regains inode_lock before it finally moves
  * inode from superblock lists we are OK.
  *
  * If `older_than_this' is non-zero then only flush inodes which have a
  * flushtime older than *older_than_this.
  *
  * If `bdi' is non-zero then we will scan the first inode against each
  * superblock until we find the matching ones.  One group will be the dirty
  * inodes against a filesystem.  Then when we hit the dummy blockdev superblock,
  * sync_sb_inodes will seekout the blockdev which matches `bdi'.  Maybe not
  * super-efficient but we're about to do a ton of I/O...
  */
 void
 writeback_inodes(struct writeback_control *wbc)
 {
 	struct super_block *sb;
 	might_sleep();
 	spin_lock(&sb_lock);
 restart:
 	sb = sb_entry(super_blocks.prev);
 	for (; sb != sb_entry(&super_blocks); sb = sb_entry(sb->s_list.prev)) {
 		if (!list_empty(&sb->s_dirty) || !list_empty(&sb->s_io)) {
 			/* we're making our own get_super here */
 			sb->s_count++;
 			spin_unlock(&sb_lock);
 			/*
 			 * If we can't get the readlock, there's no sense in
 			 * waiting around, most of the time the FS is going to
 			 * be unmounted by the time it is released.
 			 */
 			if (down_read_trylock(&sb->s_umount)) {
 				if (sb->s_root) {
 					spin_lock(&inode_lock);
 					sync_sb_inodes(sb, wbc);
 					spin_unlock(&inode_lock);
 				}
 				up_read(&sb->s_umount);
 			}
 			spin_lock(&sb_lock);
 			if (__put_super_and_need_restart(sb))
 				goto restart;
 		}
 		if (wbc->nr_to_write <= 0)
 			break;
 	}
 	spin_unlock(&sb_lock);
 }
 /*
  * writeback and wait upon the filesystem's dirty inodes.  The caller will
  * do this in two passes - one to write, and one to wait.  WB_SYNC_HOLD is
  * used to park the written inodes on sb->s_dirty for the wait pass.
  *
  * A finite limit is set on the number of pages which will be written.
  * To prevent infinite livelock of sys_sync().
  *
  * We add in the number of potentially dirty inodes, because each inode write
  * can dirty pagecache in the underlying blockdev.
  */
 void sync_inodes_sb(struct super_block *sb, int wait)
 {
 	struct writeback_control wbc = {
 		.sync_mode	= wait ? WB_SYNC_ALL : WB_SYNC_HOLD,
 	};
 	unsigned long nr_dirty = read_page_state(nr_dirty);
 	unsigned long nr_unstable = read_page_state(nr_unstable);
 	wbc.nr_to_write = nr_dirty + nr_unstable +
 			(inodes_stat.nr_inodes - inodes_stat.nr_unused) +
 			nr_dirty + nr_unstable;
 	wbc.nr_to_write += wbc.nr_to_write / 2;		/* Bit more for luck */
 	spin_lock(&inode_lock);
 	sync_sb_inodes(sb, &wbc);
 	spin_unlock(&inode_lock);
 }
 /*
  * Rather lame livelock avoidance.
  */
 static void set_sb_syncing(int val)
 {
 	struct super_block *sb;
 	spin_lock(&sb_lock);
 	sb = sb_entry(super_blocks.prev);
 	for (; sb != sb_entry(&super_blocks); sb = sb_entry(sb->s_list.prev)) {
 		sb->s_syncing = val;
 	}
 	spin_unlock(&sb_lock);
 }
 /**
  * sync_inodes - writes all inodes to disk
  * @wait: wait for completion
  *
  * sync_inodes() goes through each super block's dirty inode list, writes the
  * inodes out, waits on the writeout and puts the inodes back on the normal
  * list.
  *
  * This is for sys_sync().  fsync_dev() uses the same algorithm.  The subtle
  * part of the sync functions is that the blockdev "superblock" is processed
  * last.  This is because the write_inode() function of a typical fs will
  * perform no I/O, but will mark buffers in the blockdev mapping as dirty.
  * What we want to do is to perform all that dirtying first, and then write
  * back all those inode blocks via the blockdev mapping in one sweep.  So the
  * additional (somewhat redundant) sync_blockdev() calls here are to make
  * sure that really happens.  Because if we call sync_inodes_sb(wait=1) with
  * outstanding dirty inodes, the writeback goes block-at-a-time within the
  * filesystem's write_inode().  This is extremely slow.
  */
 static void __sync_inodes(int wait)
 {
 	struct super_block *sb;
 	spin_lock(&sb_lock);
 restart:
 	list_for_each_entry(sb, &super_blocks, s_list) {
 		if (sb->s_syncing)
 			continue;
 		sb->s_syncing = 1;
 		sb->s_count++;
 		spin_unlock(&sb_lock);
 		down_read(&sb->s_umount);
 		if (sb->s_root) {
 			sync_inodes_sb(sb, wait);
 			sync_blockdev(sb->s_bdev);
 		}
 		up_read(&sb->s_umount);
 		spin_lock(&sb_lock);
 		if (__put_super_and_need_restart(sb))
 			goto restart;
 	}
 	spin_unlock(&sb_lock);
 }
 void sync_inodes(int wait)
 {
 	set_sb_syncing(0);
 	__sync_inodes(0);
 	if (wait) {
 		set_sb_syncing(0);
 		__sync_inodes(1);
 	}
 }
 /**
  * write_inode_now	-	write an inode to disk
  * @inode: inode to write to disk
  * @sync: whether the write should be synchronous or not
  *
  * This function commits an inode to disk immediately if it is dirty. This is
  * primarily needed by knfsd.
  *
  * The caller must either have a ref on the inode or must have set I_WILL_FREE.
  */
 int write_inode_now(struct inode *inode, int sync)
 {
 	int ret;
 	struct writeback_control wbc = {
 		.nr_to_write = LONG_MAX,
 		.sync_mode = WB_SYNC_ALL,
 	};
 	if (!mapping_cap_writeback_dirty(inode->i_mapping))
 		wbc.nr_to_write = 0;
 	might_sleep();
 	spin_lock(&inode_lock);
 	ret = __writeback_single_inode(inode, &wbc);
 	spin_unlock(&inode_lock);
 	if (sync)
 		wait_on_inode(inode);
 	return ret;
 }
 EXPORT_SYMBOL(write_inode_now);
 /**
  * sync_inode - write an inode and its pages to disk.
  * @inode: the inode to sync
  * @wbc: controls the writeback mode
  *
  * sync_inode() will write an inode and its pages to disk.  It will also
  * correctly update the inode on its superblock's dirty inode lists and will
  * update inode->i_state.
  *
  * The caller must have a ref on the inode.
  */
 int sync_inode(struct inode *inode, struct writeback_control *wbc)
 {
 	int ret;
 	spin_lock(&inode_lock);
 	ret = __writeback_single_inode(inode, wbc);
 	spin_unlock(&inode_lock);
 	return ret;
 }
 EXPORT_SYMBOL(sync_inode);
 /**
  * generic_osync_inode - flush all dirty data for a given inode to disk
  * @inode: inode to write
  * @mapping: the address_space that should be flushed
  * @what:  what to write and wait upon
  *
  * This can be called by file_write functions for files which have the
  * O_SYNC flag set, to flush dirty writes to disk.
  *
  * @what is a bitmask, specifying which part of the inode's data should be
- * written and waited upon:
+ * written and waited upon.
  *
  *    OSYNC_DATA:     i_mapping's dirty data
  *    OSYNC_METADATA: the buffers at i_mapping->private_list
  *    OSYNC_INODE:    the inode itself
  */
 int generic_osync_inode(struct inode *inode, struct address_space *mapping, int what)
 {
 	int err = 0;
 	int need_write_inode_now = 0;
 	int err2;
 	current->flags |= PF_SYNCWRITE;
 	if (what & OSYNC_DATA)
 		err = filemap_fdatawrite(mapping);
 	if (what & (OSYNC_METADATA|OSYNC_DATA)) {
 		err2 = sync_mapping_buffers(mapping);
 		if (!err)
 			err = err2;
 	}
 	if (what & OSYNC_DATA) {
 		err2 = filemap_fdatawait(mapping);
 		if (!err)
 			err = err2;
 	}
 	current->flags &= ~PF_SYNCWRITE;
 	spin_lock(&inode_lock);
 	if ((inode->i_state & I_DIRTY) &&
 	    ((what & OSYNC_INODE) || (inode->i_state & I_DIRTY_DATASYNC)))
 		need_write_inode_now = 1;
 	spin_unlock(&inode_lock);
 	if (need_write_inode_now) {
 		err2 = write_inode_now(inode, 1);
 		if (!err)
 			err = err2;
 	}
 	else
 		wait_on_inode(inode);
 	return err;
 }
 EXPORT_SYMBOL(generic_osync_inode);
 /**
  * writeback_acquire: attempt to get exclusive writeback access to a device
  * @bdi: the device's backing_dev_info structure
  *
  * It is a waste of resources to have more than one pdflush thread blocked on
  * a single request queue.  Exclusion at the request_queue level is obtained
  * via a flag in the request_queue's backing_dev_info.state.
  *
  * Non-request_queue-backed address_spaces will share default_backing_dev_info,
  * unless they implement their own.  Which is somewhat inefficient, as this
  * may prevent concurrent writeback against multiple devices.
  */
 int writeback_acquire(struct backing_dev_info *bdi)
 {
 	return !test_and_set_bit(BDI_pdflush, &bdi->state);
 }
 /**
  * writeback_in_progress: determine whether there is writeback in progress
- *                        against a backing device.
  * @bdi: the device's backing_dev_info structure.
+ *
+ * Determine whether there is writeback in progress against a backing device.
  */
 int writeback_in_progress(struct backing_dev_info *bdi)
 {
 	return test_bit(BDI_pdflush, &bdi->state);
 }
 /**
  * writeback_release: relinquish exclusive writeback access against a device.
  * @bdi: the device's backing_dev_info structure
  */
 void writeback_release(struct backing_dev_info *bdi)
 {
 	BUG_ON(!writeback_in_progress(bdi));
 	clear_bit(BDI_pdflush, &bdi->state);
 }

include/linux/kernel.h

Diff comments View file @ b8887e6

 #ifndef _LINUX_KERNEL_H
 #define _LINUX_KERNEL_H
 /*
  * 'kernel.h' contains some often-used function prototypes etc
  */
 #ifdef __KERNEL__
 #include <stdarg.h>
 #include <linux/linkage.h>
 #include <linux/stddef.h>
 #include <linux/types.h>
 #include <linux/compiler.h>
 #include <linux/bitops.h>
 #include <asm/byteorder.h>
 #include <asm/bug.h>
 extern const char linux_banner[];
 #define INT_MAX		((int)(~0U>>1))
 #define INT_MIN		(-INT_MAX - 1)
 #define UINT_MAX	(~0U)
 #define LONG_MAX	((long)(~0UL>>1))
 #define LONG_MIN	(-LONG_MAX - 1)
 #define ULONG_MAX	(~0UL)
 #define STACK_MAGIC	0xdeadbeef
 #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
 #define ALIGN(x,a) (((x)+(a)-1)&~((a)-1))
 #define	KERN_EMERG	"<0>"	/* system is unusable			*/
 #define	KERN_ALERT	"<1>"	/* action must be taken immediately	*/
 #define	KERN_CRIT	"<2>"	/* critical conditions			*/
 #define	KERN_ERR	"<3>"	/* error conditions			*/
 #define	KERN_WARNING	"<4>"	/* warning conditions			*/
 #define	KERN_NOTICE	"<5>"	/* normal but significant condition	*/
 #define	KERN_INFO	"<6>"	/* informational			*/
 #define	KERN_DEBUG	"<7>"	/* debug-level messages			*/
 extern int console_printk[];
 #define console_loglevel (console_printk[0])
 #define default_message_loglevel (console_printk[1])
 #define minimum_console_loglevel (console_printk[2])
 #define default_console_loglevel (console_printk[3])
 struct completion;
 /**
  * might_sleep - annotation for functions that can sleep
  *
  * this macro will print a stack trace if it is executed in an atomic
  * context (spinlock, irq-handler, ...).
  *
  * This is a useful debugging help to be able to catch problems early and not
  * be biten later when the calling function happens to sleep when it is not
  * supposed to.
  */
 #ifdef CONFIG_PREEMPT_VOLUNTARY
 extern int cond_resched(void);
 # define might_resched() cond_resched()
 #else
 # define might_resched() do { } while (0)
 #endif
 #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP
   void __might_sleep(char *file, int line);
 # define might_sleep() \
 	do { __might_sleep(__FILE__, __LINE__); might_resched(); } while (0)
 #else
 # define might_sleep() do { might_resched(); } while (0)
 #endif
 #define might_sleep_if(cond) do { if (unlikely(cond)) might_sleep(); } while (0)
 #define abs(x) ({				\
 		int __x = (x);			\
 		(__x < 0) ? -__x : __x;		\
 	})
 #define labs(x) ({				\
 		long __x = (x);			\
 		(__x < 0) ? -__x : __x;		\
 	})
 extern struct notifier_block *panic_notifier_list;
 extern long (*panic_blink)(long time);
 NORET_TYPE void panic(const char * fmt, ...)
 	__attribute__ ((NORET_AND format (printf, 1, 2)));
 fastcall NORET_TYPE void do_exit(long error_code)
 	ATTRIB_NORET;
 NORET_TYPE void complete_and_exit(struct completion *, long)
 	ATTRIB_NORET;
 extern unsigned long simple_strtoul(const char *,char **,unsigned int);
 extern long simple_strtol(const char *,char **,unsigned int);
 extern unsigned long long simple_strtoull(const char *,char **,unsigned int);
 extern long long simple_strtoll(const char *,char **,unsigned int);
 extern int sprintf(char * buf, const char * fmt, ...)
 	__attribute__ ((format (printf, 2, 3)));
 extern int vsprintf(char *buf, const char *, va_list)
 	__attribute__ ((format (printf, 2, 0)));
 extern int snprintf(char * buf, size_t size, const char * fmt, ...)
 	__attribute__ ((format (printf, 3, 4)));
 extern int vsnprintf(char *buf, size_t size, const char *fmt, va_list args)
 	__attribute__ ((format (printf, 3, 0)));
 extern int scnprintf(char * buf, size_t size, const char * fmt, ...)
 	__attribute__ ((format (printf, 3, 4)));
 extern int vscnprintf(char *buf, size_t size, const char *fmt, va_list args)
 	__attribute__ ((format (printf, 3, 0)));
 extern int sscanf(const char *, const char *, ...)
 	__attribute__ ((format (scanf, 2, 3)));
 extern int vsscanf(const char *, const char *, va_list)
 	__attribute__ ((format (scanf, 2, 0)));
 extern int get_option(char **str, int *pint);
 extern char *get_options(const char *str, int nints, int *ints);
 extern unsigned long long memparse(char *ptr, char **retptr);
 extern int __kernel_text_address(unsigned long addr);
 extern int kernel_text_address(unsigned long addr);
 extern int session_of_pgrp(int pgrp);
 #ifdef CONFIG_PRINTK
 asmlinkage int vprintk(const char *fmt, va_list args)
 	__attribute__ ((format (printf, 1, 0)));
 asmlinkage int printk(const char * fmt, ...)
 	__attribute__ ((format (printf, 1, 2)));
 #else
 static inline int vprintk(const char *s, va_list args)
 	__attribute__ ((format (printf, 1, 0)));
 static inline int vprintk(const char *s, va_list args) { return 0; }
 static inline int printk(const char *s, ...)
 	__attribute__ ((format (printf, 1, 2)));
 static inline int printk(const char *s, ...) { return 0; }
 #endif
 unsigned long int_sqrt(unsigned long);
 static inline int __attribute_pure__ long_log2(unsigned long x)
 {
 	int r = 0;
 	for (x >>= 1; x > 0; x >>= 1)
 		r++;
 	return r;
 }
 static inline unsigned long __attribute_const__ roundup_pow_of_two(unsigned long x)
 {
 	return (1UL << fls(x - 1));
 }
 extern int printk_ratelimit(void);
 extern int __printk_ratelimit(int ratelimit_jiffies, int ratelimit_burst);
 static inline void console_silent(void)
 {
 	console_loglevel = 0;
 }
 static inline void console_verbose(void)
 {
 	if (console_loglevel)
 		console_loglevel = 15;
 }
 extern void bust_spinlocks(int yes);
 extern int oops_in_progress;		/* If set, an oops, panic(), BUG() or die() is in progress */
 extern int panic_timeout;
 extern int panic_on_oops;
 extern int tainted;
 extern const char *print_tainted(void);
 extern void add_taint(unsigned);
 /* Values used for system_state */
 extern enum system_states {
 	SYSTEM_BOOTING,
 	SYSTEM_RUNNING,
 	SYSTEM_HALT,
 	SYSTEM_POWER_OFF,
 	SYSTEM_RESTART,
 } system_state;
 #define TAINT_PROPRIETARY_MODULE	(1<<0)
 #define TAINT_FORCED_MODULE		(1<<1)
 #define TAINT_UNSAFE_SMP		(1<<2)
 #define TAINT_FORCED_RMMOD		(1<<3)
 #define TAINT_MACHINE_CHECK		(1<<4)
 #define TAINT_BAD_PAGE			(1<<5)
 extern void dump_stack(void);
 #ifdef DEBUG
 #define pr_debug(fmt,arg...) \
 	printk(KERN_DEBUG fmt,##arg)
 #else
 #define pr_debug(fmt,arg...) \
 	do { } while (0)
 #endif
 #define pr_info(fmt,arg...) \
 	printk(KERN_INFO fmt,##arg)
 /*
  *      Display an IP address in readable format.
  */
 #define NIPQUAD(addr) \
 	((unsigned char *)&addr)[0], \
 	((unsigned char *)&addr)[1], \
 	((unsigned char *)&addr)[2], \
 	((unsigned char *)&addr)[3]
 #define NIP6(addr) \
 	ntohs((addr).s6_addr16[0]), \
 	ntohs((addr).s6_addr16[1]), \
 	ntohs((addr).s6_addr16[2]), \
 	ntohs((addr).s6_addr16[3]), \
 	ntohs((addr).s6_addr16[4]), \
 	ntohs((addr).s6_addr16[5]), \
 	ntohs((addr).s6_addr16[6]), \
 	ntohs((addr).s6_addr16[7])
 #if defined(__LITTLE_ENDIAN)
 #define HIPQUAD(addr) \
 	((unsigned char *)&addr)[3], \
 	((unsigned char *)&addr)[2], \
 	((unsigned char *)&addr)[1], \
 	((unsigned char *)&addr)[0]
 #elif defined(__BIG_ENDIAN)
 #define HIPQUAD	NIPQUAD
 #else
 #error "Please fix asm/byteorder.h"
 #endif /* __LITTLE_ENDIAN */
 /*
  * min()/max() macros that also do
  * strict type-checking.. See the
  * "unnecessary" pointer comparison.
  */
 #define min(x,y) ({ \
 	typeof(x) _x = (x);	\
 	typeof(y) _y = (y);	\
 	(void) (&_x == &_y);		\
 	_x < _y ? _x : _y; })
 #define max(x,y) ({ \
 	typeof(x) _x = (x);	\
 	typeof(y) _y = (y);	\
 	(void) (&_x == &_y);		\
 	_x > _y ? _x : _y; })
 /*
  * ..and if you can't take the strict
  * types, you can specify one yourself.
  *
  * Or not use min/max at all, of course.
  */
 #define min_t(type,x,y) \
 	({ type __x = (x); type __y = (y); __x < __y ? __x: __y; })
 #define max_t(type,x,y) \
 	({ type __x = (x); type __y = (y); __x > __y ? __x: __y; })
 /**
  * container_of - cast a member of a structure out to the containing structure
- *
  * @ptr:	the pointer to the member.
  * @type:	the type of the container struct this is embedded in.
  * @member:	the name of the member within the struct.
  *
  */
 #define container_of(ptr, type, member) ({			\
         const typeof( ((type *)0)->member ) *__mptr = (ptr);	\
         (type *)( (char *)__mptr - offsetof(type,member) );})
 /*
  * Check at compile time that something is of a particular type.
  * Always evaluates to 1 so you may use it easily in comparisons.
  */
 #define typecheck(type,x) \
 ({	type __dummy; \
 	typeof(x) __dummy2; \
 	(void)(&__dummy == &__dummy2); \
 	1; \
 })
 #endif /* __KERNEL__ */
 #define SI_LOAD_SHIFT	16
 struct sysinfo {
 	long uptime;			/* Seconds since boot */
 	unsigned long loads[3];		/* 1, 5, and 15 minute load averages */
 	unsigned long totalram;		/* Total usable main memory size */
 	unsigned long freeram;		/* Available memory size */
 	unsigned long sharedram;	/* Amount of shared memory */
 	unsigned long bufferram;	/* Memory used by buffers */
 	unsigned long totalswap;	/* Total swap space size */
 	unsigned long freeswap;		/* swap space still available */
 	unsigned short procs;		/* Number of current processes */
 	unsigned short pad;		/* explicit padding for m68k */
 	unsigned long totalhigh;	/* Total high memory size */
 	unsigned long freehigh;		/* Available high memory size */
 	unsigned int mem_unit;		/* Memory unit size in bytes */
 	char _f[20-2*sizeof(long)-sizeof(int)];	/* Padding: libc5 uses this.. */
 };
 /* Force a compilation error if condition is true */
 #define BUILD_BUG_ON(condition) ((void)sizeof(char[1 - 2*!!(condition)]))
 #ifdef CONFIG_SYSCTL
 extern int randomize_va_space;
 #else
 #define randomize_va_space 1
 #endif
 /* Trap pasters of __FUNCTION__ at compile-time */
 #if __GNUC__ > 2 || __GNUC_MINOR__ >= 95
 #define __FUNCTION__ (__func__)
 #endif
 #endif

kernel/sys.c

Diff comments View file @ b8887e6

1	/*	1	/*
2	* linux/kernel/sys.c	2	* linux/kernel/sys.c
3	*	3	*
4	* Copyright (C) 1991, 1992 Linus Torvalds	4	* Copyright (C) 1991, 1992 Linus Torvalds
5	*/	5	*/
6		6
7	#include <linux/config.h>	7	#include <linux/config.h>
8	#include <linux/module.h>	8	#include <linux/module.h>
9	#include <linux/mm.h>	9	#include <linux/mm.h>
10	#include <linux/utsname.h>	10	#include <linux/utsname.h>
11	#include <linux/mman.h>	11	#include <linux/mman.h>
12	#include <linux/smp_lock.h>	12	#include <linux/smp_lock.h>
13	#include <linux/notifier.h>	13	#include <linux/notifier.h>
14	#include <linux/reboot.h>	14	#include <linux/reboot.h>
15	#include <linux/prctl.h>	15	#include <linux/prctl.h>
16	#include <linux/init.h>	16	#include <linux/init.h>
17	#include <linux/highuid.h>	17	#include <linux/highuid.h>
18	#include <linux/fs.h>	18	#include <linux/fs.h>
19	#include <linux/kernel.h>	19	#include <linux/kernel.h>
20	#include <linux/kexec.h>	20	#include <linux/kexec.h>
21	#include <linux/workqueue.h>	21	#include <linux/workqueue.h>
22	#include <linux/device.h>	22	#include <linux/device.h>
23	#include <linux/key.h>	23	#include <linux/key.h>
24	#include <linux/times.h>	24	#include <linux/times.h>
25	#include <linux/posix-timers.h>	25	#include <linux/posix-timers.h>
26	#include <linux/security.h>	26	#include <linux/security.h>
27	#include <linux/dcookies.h>	27	#include <linux/dcookies.h>
28	#include <linux/suspend.h>	28	#include <linux/suspend.h>
29	#include <linux/tty.h>	29	#include <linux/tty.h>
30	#include <linux/signal.h>	30	#include <linux/signal.h>
31	#include <linux/cn_proc.h>	31	#include <linux/cn_proc.h>
32		32
33	#include <linux/compat.h>	33	#include <linux/compat.h>
34	#include <linux/syscalls.h>	34	#include <linux/syscalls.h>
35		35
36	#include <asm/uaccess.h>	36	#include <asm/uaccess.h>
37	#include <asm/io.h>	37	#include <asm/io.h>
38	#include <asm/unistd.h>	38	#include <asm/unistd.h>
39		39
40	#ifndef SET_UNALIGN_CTL	40	#ifndef SET_UNALIGN_CTL
41	# define SET_UNALIGN_CTL(a,b) (-EINVAL)	41	# define SET_UNALIGN_CTL(a,b) (-EINVAL)
42	#endif	42	#endif
43	#ifndef GET_UNALIGN_CTL	43	#ifndef GET_UNALIGN_CTL
44	# define GET_UNALIGN_CTL(a,b) (-EINVAL)	44	# define GET_UNALIGN_CTL(a,b) (-EINVAL)
45	#endif	45	#endif
46	#ifndef SET_FPEMU_CTL	46	#ifndef SET_FPEMU_CTL
47	# define SET_FPEMU_CTL(a,b) (-EINVAL)	47	# define SET_FPEMU_CTL(a,b) (-EINVAL)
48	#endif	48	#endif
49	#ifndef GET_FPEMU_CTL	49	#ifndef GET_FPEMU_CTL
50	# define GET_FPEMU_CTL(a,b) (-EINVAL)	50	# define GET_FPEMU_CTL(a,b) (-EINVAL)
51	#endif	51	#endif
52	#ifndef SET_FPEXC_CTL	52	#ifndef SET_FPEXC_CTL
53	# define SET_FPEXC_CTL(a,b) (-EINVAL)	53	# define SET_FPEXC_CTL(a,b) (-EINVAL)
54	#endif	54	#endif
55	#ifndef GET_FPEXC_CTL	55	#ifndef GET_FPEXC_CTL
56	# define GET_FPEXC_CTL(a,b) (-EINVAL)	56	# define GET_FPEXC_CTL(a,b) (-EINVAL)
57	#endif	57	#endif
58		58
59	/*	59	/*
60	* this is where the system-wide overflow UID and GID are defined, for	60	* this is where the system-wide overflow UID and GID are defined, for
61	* architectures that now have 32-bit UID/GID but didn't in the past	61	* architectures that now have 32-bit UID/GID but didn't in the past
62	*/	62	*/
63		63
64	int overflowuid = DEFAULT_OVERFLOWUID;	64	int overflowuid = DEFAULT_OVERFLOWUID;
65	int overflowgid = DEFAULT_OVERFLOWGID;	65	int overflowgid = DEFAULT_OVERFLOWGID;
66		66
67	#ifdef CONFIG_UID16	67	#ifdef CONFIG_UID16
68	EXPORT_SYMBOL(overflowuid);	68	EXPORT_SYMBOL(overflowuid);
69	EXPORT_SYMBOL(overflowgid);	69	EXPORT_SYMBOL(overflowgid);
70	#endif	70	#endif
71		71
72	/*	72	/*
73	* the same as above, but for filesystems which can only store a 16-bit	73	* the same as above, but for filesystems which can only store a 16-bit
74	* UID and GID. as such, this is needed on all architectures	74	* UID and GID. as such, this is needed on all architectures
75	*/	75	*/
76		76
77	int fs_overflowuid = DEFAULT_FS_OVERFLOWUID;	77	int fs_overflowuid = DEFAULT_FS_OVERFLOWUID;
78	int fs_overflowgid = DEFAULT_FS_OVERFLOWUID;	78	int fs_overflowgid = DEFAULT_FS_OVERFLOWUID;
79		79
80	EXPORT_SYMBOL(fs_overflowuid);	80	EXPORT_SYMBOL(fs_overflowuid);
81	EXPORT_SYMBOL(fs_overflowgid);	81	EXPORT_SYMBOL(fs_overflowgid);
82		82
83	/*	83	/*
84	* this indicates whether you can reboot with ctrl-alt-del: the default is yes	84	* this indicates whether you can reboot with ctrl-alt-del: the default is yes
85	*/	85	*/
86		86
87	int C_A_D = 1;	87	int C_A_D = 1;
88	int cad_pid = 1;	88	int cad_pid = 1;
89		89
90	/*	90	/*
91	* Notifier list for kernel code which wants to be called	91	* Notifier list for kernel code which wants to be called
92	* at shutdown. This is used to stop any idling DMA operations	92	* at shutdown. This is used to stop any idling DMA operations
93	* and the like.	93	* and the like.
94	*/	94	*/
95		95
96	static struct notifier_block *reboot_notifier_list;	96	static struct notifier_block *reboot_notifier_list;
97	static DEFINE_RWLOCK(notifier_lock);	97	static DEFINE_RWLOCK(notifier_lock);
98		98
99	/**	99	/**
100	* notifier_chain_register - Add notifier to a notifier chain	100	* notifier_chain_register - Add notifier to a notifier chain
101	* @list: Pointer to root list pointer	101	* @list: Pointer to root list pointer
102	* @n: New entry in notifier chain	102	* @n: New entry in notifier chain
103	*	103	*
104	* Adds a notifier to a notifier chain.	104	* Adds a notifier to a notifier chain.
105	*	105	*
106	* Currently always returns zero.	106	* Currently always returns zero.
107	*/	107	*/
108		108
109	int notifier_chain_register(struct notifier_block *list, struct notifier_block n)	109	int notifier_chain_register(struct notifier_block *list, struct notifier_block n)
110	{	110	{
111	write_lock(&notifier_lock);	111	write_lock(&notifier_lock);
112	while(*list)	112	while(*list)
113	{	113	{
114	if(n->priority > (*list)->priority)	114	if(n->priority > (*list)->priority)
115	break;	115	break;
116	list= &((*list)->next);	116	list= &((*list)->next);
117	}	117	}
118	n->next = *list;	118	n->next = *list;
119	*list=n;	119	*list=n;
120	write_unlock(&notifier_lock);	120	write_unlock(&notifier_lock);
121	return 0;	121	return 0;
122	}	122	}
123		123
124	EXPORT_SYMBOL(notifier_chain_register);	124	EXPORT_SYMBOL(notifier_chain_register);
125		125
126	/**	126	/**
127	* notifier_chain_unregister - Remove notifier from a notifier chain	127	* notifier_chain_unregister - Remove notifier from a notifier chain
128	* @nl: Pointer to root list pointer	128	* @nl: Pointer to root list pointer
129	* @n: New entry in notifier chain	129	* @n: New entry in notifier chain
130	*	130	*
131	* Removes a notifier from a notifier chain.	131	* Removes a notifier from a notifier chain.
132	*	132	*
133	* Returns zero on success, or %-ENOENT on failure.	133	* Returns zero on success, or %-ENOENT on failure.
134	*/	134	*/
135		135
136	int notifier_chain_unregister(struct notifier_block *nl, struct notifier_block n)	136	int notifier_chain_unregister(struct notifier_block *nl, struct notifier_block n)
137	{	137	{
138	write_lock(&notifier_lock);	138	write_lock(&notifier_lock);
139	while((*nl)!=NULL)	139	while((*nl)!=NULL)
140	{	140	{
141	if((*nl)==n)	141	if((*nl)==n)
142	{	142	{
143	*nl=n->next;	143	*nl=n->next;
144	write_unlock(&notifier_lock);	144	write_unlock(&notifier_lock);
145	return 0;	145	return 0;
146	}	146	}
147	nl=&((*nl)->next);	147	nl=&((*nl)->next);
148	}	148	}
149	write_unlock(&notifier_lock);	149	write_unlock(&notifier_lock);
150	return -ENOENT;	150	return -ENOENT;
151	}	151	}
152		152
153	EXPORT_SYMBOL(notifier_chain_unregister);	153	EXPORT_SYMBOL(notifier_chain_unregister);
154		154
155	/**	155	/**
156	* notifier_call_chain - Call functions in a notifier chain	156	* notifier_call_chain - Call functions in a notifier chain
157	* @n: Pointer to root pointer of notifier chain	157	* @n: Pointer to root pointer of notifier chain
158	* @val: Value passed unmodified to notifier function	158	* @val: Value passed unmodified to notifier function
159	* @v: Pointer passed unmodified to notifier function	159	* @v: Pointer passed unmodified to notifier function
160	*	160	*
161	* Calls each function in a notifier chain in turn.	161	* Calls each function in a notifier chain in turn.
162	*	162	*
163	* If the return value of the notifier can be and'd	163	* If the return value of the notifier can be and'd
164	* with %NOTIFY_STOP_MASK, then notifier_call_chain	164	* with %NOTIFY_STOP_MASK, then notifier_call_chain
165	* will return immediately, with the return value of	165	* will return immediately, with the return value of
166	* the notifier function which halted execution.	166	* the notifier function which halted execution.
167	* Otherwise, the return value is the return value	167	* Otherwise, the return value is the return value
168	* of the last notifier function called.	168	* of the last notifier function called.
169	*/	169	*/
170		170
171	int notifier_call_chain(struct notifier_block *n, unsigned long val, void v)	171	int notifier_call_chain(struct notifier_block *n, unsigned long val, void v)
172	{	172	{
173	int ret=NOTIFY_DONE;	173	int ret=NOTIFY_DONE;
174	struct notifier_block nb = n;	174	struct notifier_block nb = n;
175		175
176	while(nb)	176	while(nb)
177	{	177	{
178	ret=nb->notifier_call(nb,val,v);	178	ret=nb->notifier_call(nb,val,v);
179	if(ret&NOTIFY_STOP_MASK)	179	if(ret&NOTIFY_STOP_MASK)
180	{	180	{
181	return ret;	181	return ret;
182	}	182	}
183	nb=nb->next;	183	nb=nb->next;
184	}	184	}
185	return ret;	185	return ret;
186	}	186	}
187		187
188	EXPORT_SYMBOL(notifier_call_chain);	188	EXPORT_SYMBOL(notifier_call_chain);
189		189
190	/**	190	/**
191	* register_reboot_notifier - Register function to be called at reboot time	191	* register_reboot_notifier - Register function to be called at reboot time
192	* @nb: Info about notifier function to be called	192	* @nb: Info about notifier function to be called
193	*	193	*
194	* Registers a function with the list of functions	194	* Registers a function with the list of functions
195	* to be called at reboot time.	195	* to be called at reboot time.
196	*	196	*
197	* Currently always returns zero, as notifier_chain_register	197	* Currently always returns zero, as notifier_chain_register
198	* always returns zero.	198	* always returns zero.
199	*/	199	*/
200		200
201	int register_reboot_notifier(struct notifier_block * nb)	201	int register_reboot_notifier(struct notifier_block * nb)
202	{	202	{
203	return notifier_chain_register(&reboot_notifier_list, nb);	203	return notifier_chain_register(&reboot_notifier_list, nb);
204	}	204	}
205		205
206	EXPORT_SYMBOL(register_reboot_notifier);	206	EXPORT_SYMBOL(register_reboot_notifier);
207		207
208	/**	208	/**
209	* unregister_reboot_notifier - Unregister previously registered reboot notifier	209	* unregister_reboot_notifier - Unregister previously registered reboot notifier
210	* @nb: Hook to be unregistered	210	* @nb: Hook to be unregistered
211	*	211	*
212	* Unregisters a previously registered reboot	212	* Unregisters a previously registered reboot
213	* notifier function.	213	* notifier function.
214	*	214	*
215	* Returns zero on success, or %-ENOENT on failure.	215	* Returns zero on success, or %-ENOENT on failure.
216	*/	216	*/
217		217
218	int unregister_reboot_notifier(struct notifier_block * nb)	218	int unregister_reboot_notifier(struct notifier_block * nb)
219	{	219	{
220	return notifier_chain_unregister(&reboot_notifier_list, nb);	220	return notifier_chain_unregister(&reboot_notifier_list, nb);
221	}	221	}
222		222
223	EXPORT_SYMBOL(unregister_reboot_notifier);	223	EXPORT_SYMBOL(unregister_reboot_notifier);
224		224
225	static int set_one_prio(struct task_struct *p, int niceval, int error)	225	static int set_one_prio(struct task_struct *p, int niceval, int error)
226	{	226	{
227	int no_nice;	227	int no_nice;
228		228
229	if (p->uid != current->euid &&	229	if (p->uid != current->euid &&
230	p->euid != current->euid && !capable(CAP_SYS_NICE)) {	230	p->euid != current->euid && !capable(CAP_SYS_NICE)) {
231	error = -EPERM;	231	error = -EPERM;
232	goto out;	232	goto out;
233	}	233	}
234	if (niceval < task_nice(p) && !can_nice(p, niceval)) {	234	if (niceval < task_nice(p) && !can_nice(p, niceval)) {
235	error = -EACCES;	235	error = -EACCES;
236	goto out;	236	goto out;
237	}	237	}
238	no_nice = security_task_setnice(p, niceval);	238	no_nice = security_task_setnice(p, niceval);
239	if (no_nice) {	239	if (no_nice) {
240	error = no_nice;	240	error = no_nice;
241	goto out;	241	goto out;
242	}	242	}
243	if (error == -ESRCH)	243	if (error == -ESRCH)
244	error = 0;	244	error = 0;
245	set_user_nice(p, niceval);	245	set_user_nice(p, niceval);
246	out:	246	out:
247	return error;	247	return error;
248	}	248	}
249		249
250	asmlinkage long sys_setpriority(int which, int who, int niceval)	250	asmlinkage long sys_setpriority(int which, int who, int niceval)
251	{	251	{
252	struct task_struct g, p;	252	struct task_struct g, p;
253	struct user_struct *user;	253	struct user_struct *user;
254	int error = -EINVAL;	254	int error = -EINVAL;
255		255
256	if (which > 2 \|\| which < 0)	256	if (which > 2 \|\| which < 0)
257	goto out;	257	goto out;
258		258
259	/* normalize: avoid signed division (rounding problems) */	259	/* normalize: avoid signed division (rounding problems) */
260	error = -ESRCH;	260	error = -ESRCH;
261	if (niceval < -20)	261	if (niceval < -20)
262	niceval = -20;	262	niceval = -20;
263	if (niceval > 19)	263	if (niceval > 19)
264	niceval = 19;	264	niceval = 19;
265		265
266	read_lock(&tasklist_lock);	266	read_lock(&tasklist_lock);
267	switch (which) {	267	switch (which) {
268	case PRIO_PROCESS:	268	case PRIO_PROCESS:
269	if (!who)	269	if (!who)
270	who = current->pid;	270	who = current->pid;
271	p = find_task_by_pid(who);	271	p = find_task_by_pid(who);
272	if (p)	272	if (p)
273	error = set_one_prio(p, niceval, error);	273	error = set_one_prio(p, niceval, error);
274	break;	274	break;
275	case PRIO_PGRP:	275	case PRIO_PGRP:
276	if (!who)	276	if (!who)
277	who = process_group(current);	277	who = process_group(current);
278	do_each_task_pid(who, PIDTYPE_PGID, p) {	278	do_each_task_pid(who, PIDTYPE_PGID, p) {
279	error = set_one_prio(p, niceval, error);	279	error = set_one_prio(p, niceval, error);
280	} while_each_task_pid(who, PIDTYPE_PGID, p);	280	} while_each_task_pid(who, PIDTYPE_PGID, p);
281	break;	281	break;
282	case PRIO_USER:	282	case PRIO_USER:
283	user = current->user;	283	user = current->user;
284	if (!who)	284	if (!who)
285	who = current->uid;	285	who = current->uid;
286	else	286	else
287	if ((who != current->uid) && !(user = find_user(who)))	287	if ((who != current->uid) && !(user = find_user(who)))
288	goto out_unlock; /* No processes for this user */	288	goto out_unlock; /* No processes for this user */
289		289
290	do_each_thread(g, p)	290	do_each_thread(g, p)
291	if (p->uid == who)	291	if (p->uid == who)
292	error = set_one_prio(p, niceval, error);	292	error = set_one_prio(p, niceval, error);
293	while_each_thread(g, p);	293	while_each_thread(g, p);
294	if (who != current->uid)	294	if (who != current->uid)
295	free_uid(user); /* For find_user() */	295	free_uid(user); /* For find_user() */
296	break;	296	break;
297	}	297	}
298	out_unlock:	298	out_unlock:
299	read_unlock(&tasklist_lock);	299	read_unlock(&tasklist_lock);
300	out:	300	out:
301	return error;	301	return error;
302	}	302	}
303		303
304	/*	304	/*
305	* Ugh. To avoid negative return values, "getpriority()" will	305	* Ugh. To avoid negative return values, "getpriority()" will
306	* not return the normal nice-value, but a negated value that	306	* not return the normal nice-value, but a negated value that
307	* has been offset by 20 (ie it returns 40..1 instead of -20..19)	307	* has been offset by 20 (ie it returns 40..1 instead of -20..19)
308	* to stay compatible.	308	* to stay compatible.
309	*/	309	*/
310	asmlinkage long sys_getpriority(int which, int who)	310	asmlinkage long sys_getpriority(int which, int who)
311	{	311	{
312	struct task_struct g, p;	312	struct task_struct g, p;
313	struct user_struct *user;	313	struct user_struct *user;
314	long niceval, retval = -ESRCH;	314	long niceval, retval = -ESRCH;
315		315
316	if (which > 2 \|\| which < 0)	316	if (which > 2 \|\| which < 0)
317	return -EINVAL;	317	return -EINVAL;
318		318
319	read_lock(&tasklist_lock);	319	read_lock(&tasklist_lock);
320	switch (which) {	320	switch (which) {
321	case PRIO_PROCESS:	321	case PRIO_PROCESS:
322	if (!who)	322	if (!who)
323	who = current->pid;	323	who = current->pid;
324	p = find_task_by_pid(who);	324	p = find_task_by_pid(who);
325	if (p) {	325	if (p) {
326	niceval = 20 - task_nice(p);	326	niceval = 20 - task_nice(p);
327	if (niceval > retval)	327	if (niceval > retval)
328	retval = niceval;	328	retval = niceval;
329	}	329	}
330	break;	330	break;
331	case PRIO_PGRP:	331	case PRIO_PGRP:
332	if (!who)	332	if (!who)
333	who = process_group(current);	333	who = process_group(current);
334	do_each_task_pid(who, PIDTYPE_PGID, p) {	334	do_each_task_pid(who, PIDTYPE_PGID, p) {
335	niceval = 20 - task_nice(p);	335	niceval = 20 - task_nice(p);
336	if (niceval > retval)	336	if (niceval > retval)
337	retval = niceval;	337	retval = niceval;
338	} while_each_task_pid(who, PIDTYPE_PGID, p);	338	} while_each_task_pid(who, PIDTYPE_PGID, p);
339	break;	339	break;
340	case PRIO_USER:	340	case PRIO_USER:
341	user = current->user;	341	user = current->user;
342	if (!who)	342	if (!who)
343	who = current->uid;	343	who = current->uid;
344	else	344	else
345	if ((who != current->uid) && !(user = find_user(who)))	345	if ((who != current->uid) && !(user = find_user(who)))
346	goto out_unlock; /* No processes for this user */	346	goto out_unlock; /* No processes for this user */
347		347
348	do_each_thread(g, p)	348	do_each_thread(g, p)
349	if (p->uid == who) {	349	if (p->uid == who) {
350	niceval = 20 - task_nice(p);	350	niceval = 20 - task_nice(p);
351	if (niceval > retval)	351	if (niceval > retval)
352	retval = niceval;	352	retval = niceval;
353	}	353	}
354	while_each_thread(g, p);	354	while_each_thread(g, p);
355	if (who != current->uid)	355	if (who != current->uid)
356	free_uid(user); /* for find_user() */	356	free_uid(user); /* for find_user() */
357	break;	357	break;
358	}	358	}
359	out_unlock:	359	out_unlock:
360	read_unlock(&tasklist_lock);	360	read_unlock(&tasklist_lock);
361		361
362	return retval;	362	return retval;
363	}	363	}
364		364
365	/**	365	/**
366	* emergency_restart - reboot the system	366	* emergency_restart - reboot the system
367	*	367	*
368	* Without shutting down any hardware or taking any locks	368	* Without shutting down any hardware or taking any locks
369	* reboot the system. This is called when we know we are in	369	* reboot the system. This is called when we know we are in
370	* trouble so this is our best effort to reboot. This is	370	* trouble so this is our best effort to reboot. This is
371	* safe to call in interrupt context.	371	* safe to call in interrupt context.
372	*/	372	*/
373	void emergency_restart(void)	373	void emergency_restart(void)
374	{	374	{
375	machine_emergency_restart();	375	machine_emergency_restart();
376	}	376	}
377	EXPORT_SYMBOL_GPL(emergency_restart);	377	EXPORT_SYMBOL_GPL(emergency_restart);
378		378
379	void kernel_restart_prepare(char *cmd)	379	void kernel_restart_prepare(char *cmd)
380	{	380	{
381	notifier_call_chain(&reboot_notifier_list, SYS_RESTART, cmd);	381	notifier_call_chain(&reboot_notifier_list, SYS_RESTART, cmd);
382	system_state = SYSTEM_RESTART;	382	system_state = SYSTEM_RESTART;
383	device_shutdown();	383	device_shutdown();
384	}	384	}
385		385
386	/**	386	/**
387	* kernel_restart - reboot the system	387	* kernel_restart - reboot the system
388	* @cmd: pointer to buffer containing command to execute for restart	388	* @cmd: pointer to buffer containing command to execute for restart
389	* or NULL	389	* or %NULL
390	*	390	*
391	* Shutdown everything and perform a clean reboot.	391	* Shutdown everything and perform a clean reboot.
392	* This is not safe to call in interrupt context.	392	* This is not safe to call in interrupt context.
393	*/	393	*/
394	void kernel_restart(char *cmd)	394	void kernel_restart(char *cmd)
395	{	395	{
396	kernel_restart_prepare(cmd);	396	kernel_restart_prepare(cmd);
397	if (!cmd) {	397	if (!cmd) {
398	printk(KERN_EMERG "Restarting system.\n");	398	printk(KERN_EMERG "Restarting system.\n");
399	} else {	399	} else {
400	printk(KERN_EMERG "Restarting system with command '%s'.\n", cmd);	400	printk(KERN_EMERG "Restarting system with command '%s'.\n", cmd);
401	}	401	}
402	printk(".\n");	402	printk(".\n");
403	machine_restart(cmd);	403	machine_restart(cmd);
404	}	404	}
405	EXPORT_SYMBOL_GPL(kernel_restart);	405	EXPORT_SYMBOL_GPL(kernel_restart);
406		406
407	/**	407	/**
408	* kernel_kexec - reboot the system	408	* kernel_kexec - reboot the system
409	*	409	*
410	* Move into place and start executing a preloaded standalone	410	* Move into place and start executing a preloaded standalone
411	* executable. If nothing was preloaded return an error.	411	* executable. If nothing was preloaded return an error.
412	*/	412	*/
413	void kernel_kexec(void)	413	void kernel_kexec(void)
414	{	414	{
415	#ifdef CONFIG_KEXEC	415	#ifdef CONFIG_KEXEC
416	struct kimage *image;	416	struct kimage *image;
417	image = xchg(&kexec_image, 0);	417	image = xchg(&kexec_image, 0);
418	if (!image) {	418	if (!image) {
419	return;	419	return;
420	}	420	}
421	kernel_restart_prepare(NULL);	421	kernel_restart_prepare(NULL);
422	printk(KERN_EMERG "Starting new kernel\n");	422	printk(KERN_EMERG "Starting new kernel\n");
423	machine_shutdown();	423	machine_shutdown();
424	machine_kexec(image);	424	machine_kexec(image);
425	#endif	425	#endif
426	}	426	}
427	EXPORT_SYMBOL_GPL(kernel_kexec);	427	EXPORT_SYMBOL_GPL(kernel_kexec);
428		428
429	/**	429	/**
430	* kernel_halt - halt the system	430	* kernel_halt - halt the system
431	*	431	*
432	* Shutdown everything and perform a clean system halt.	432	* Shutdown everything and perform a clean system halt.
433	*/	433	*/
434	void kernel_halt_prepare(void)	434	void kernel_halt_prepare(void)
435	{	435	{
436	notifier_call_chain(&reboot_notifier_list, SYS_HALT, NULL);	436	notifier_call_chain(&reboot_notifier_list, SYS_HALT, NULL);
437	system_state = SYSTEM_HALT;	437	system_state = SYSTEM_HALT;
438	device_shutdown();	438	device_shutdown();
439	}	439	}
440	void kernel_halt(void)	440	void kernel_halt(void)
441	{	441	{
442	kernel_halt_prepare();	442	kernel_halt_prepare();
443	printk(KERN_EMERG "System halted.\n");	443	printk(KERN_EMERG "System halted.\n");
444	machine_halt();	444	machine_halt();
445	}	445	}
446	EXPORT_SYMBOL_GPL(kernel_halt);	446	EXPORT_SYMBOL_GPL(kernel_halt);
447		447
448	/**	448	/**
449	* kernel_power_off - power_off the system	449	* kernel_power_off - power_off the system
450	*	450	*
451	* Shutdown everything and perform a clean system power_off.	451	* Shutdown everything and perform a clean system power_off.
452	*/	452	*/
453	void kernel_power_off_prepare(void)	453	void kernel_power_off_prepare(void)
454	{	454	{
455	notifier_call_chain(&reboot_notifier_list, SYS_POWER_OFF, NULL);	455	notifier_call_chain(&reboot_notifier_list, SYS_POWER_OFF, NULL);
456	system_state = SYSTEM_POWER_OFF;	456	system_state = SYSTEM_POWER_OFF;
457	device_shutdown();	457	device_shutdown();
458	}	458	}
459	void kernel_power_off(void)	459	void kernel_power_off(void)
460	{	460	{
461	kernel_power_off_prepare();	461	kernel_power_off_prepare();
462	printk(KERN_EMERG "Power down.\n");	462	printk(KERN_EMERG "Power down.\n");
463	machine_power_off();	463	machine_power_off();
464	}	464	}
465	EXPORT_SYMBOL_GPL(kernel_power_off);	465	EXPORT_SYMBOL_GPL(kernel_power_off);
466		466
467	/*	467	/*
468	* Reboot system call: for obvious reasons only root may call it,	468	* Reboot system call: for obvious reasons only root may call it,
469	* and even root needs to set up some magic numbers in the registers	469	* and even root needs to set up some magic numbers in the registers
470	* so that some mistake won't make this reboot the whole machine.	470	* so that some mistake won't make this reboot the whole machine.
471	* You can also set the meaning of the ctrl-alt-del-key here.	471	* You can also set the meaning of the ctrl-alt-del-key here.
472	*	472	*
473	* reboot doesn't sync: do that yourself before calling this.	473	* reboot doesn't sync: do that yourself before calling this.
474	*/	474	*/
475	asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user * arg)	475	asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user * arg)
476	{	476	{
477	char buffer[256];	477	char buffer[256];
478		478
479	/* We only trust the superuser with rebooting the system. */	479	/* We only trust the superuser with rebooting the system. */
480	if (!capable(CAP_SYS_BOOT))	480	if (!capable(CAP_SYS_BOOT))
481	return -EPERM;	481	return -EPERM;
482		482
483	/* For safety, we require "magic" arguments. */	483	/* For safety, we require "magic" arguments. */
484	if (magic1 != LINUX_REBOOT_MAGIC1 \|\|	484	if (magic1 != LINUX_REBOOT_MAGIC1 \|\|
485	(magic2 != LINUX_REBOOT_MAGIC2 &&	485	(magic2 != LINUX_REBOOT_MAGIC2 &&
486	magic2 != LINUX_REBOOT_MAGIC2A &&	486	magic2 != LINUX_REBOOT_MAGIC2A &&
487	magic2 != LINUX_REBOOT_MAGIC2B &&	487	magic2 != LINUX_REBOOT_MAGIC2B &&
488	magic2 != LINUX_REBOOT_MAGIC2C))	488	magic2 != LINUX_REBOOT_MAGIC2C))
489	return -EINVAL;	489	return -EINVAL;
490		490
491	lock_kernel();	491	lock_kernel();
492	switch (cmd) {	492	switch (cmd) {
493	case LINUX_REBOOT_CMD_RESTART:	493	case LINUX_REBOOT_CMD_RESTART:
494	kernel_restart(NULL);	494	kernel_restart(NULL);
495	break;	495	break;
496		496
497	case LINUX_REBOOT_CMD_CAD_ON:	497	case LINUX_REBOOT_CMD_CAD_ON:
498	C_A_D = 1;	498	C_A_D = 1;
499	break;	499	break;
500		500
501	case LINUX_REBOOT_CMD_CAD_OFF:	501	case LINUX_REBOOT_CMD_CAD_OFF:
502	C_A_D = 0;	502	C_A_D = 0;
503	break;	503	break;
504		504
505	case LINUX_REBOOT_CMD_HALT:	505	case LINUX_REBOOT_CMD_HALT:
506	kernel_halt();	506	kernel_halt();
507	unlock_kernel();	507	unlock_kernel();
508	do_exit(0);	508	do_exit(0);
509	break;	509	break;
510		510
511	case LINUX_REBOOT_CMD_POWER_OFF:	511	case LINUX_REBOOT_CMD_POWER_OFF:
512	kernel_power_off();	512	kernel_power_off();
513	unlock_kernel();	513	unlock_kernel();
514	do_exit(0);	514	do_exit(0);
515	break;	515	break;
516		516
517	case LINUX_REBOOT_CMD_RESTART2:	517	case LINUX_REBOOT_CMD_RESTART2:
518	if (strncpy_from_user(&buffer[0], arg, sizeof(buffer) - 1) < 0) {	518	if (strncpy_from_user(&buffer[0], arg, sizeof(buffer) - 1) < 0) {
519	unlock_kernel();	519	unlock_kernel();
520	return -EFAULT;	520	return -EFAULT;
521	}	521	}
522	buffer[sizeof(buffer) - 1] = '\0';	522	buffer[sizeof(buffer) - 1] = '\0';
523		523
524	kernel_restart(buffer);	524	kernel_restart(buffer);
525	break;	525	break;
526		526
527	case LINUX_REBOOT_CMD_KEXEC:	527	case LINUX_REBOOT_CMD_KEXEC:
528	kernel_kexec();	528	kernel_kexec();
529	unlock_kernel();	529	unlock_kernel();
530	return -EINVAL;	530	return -EINVAL;
531		531
532	#ifdef CONFIG_SOFTWARE_SUSPEND	532	#ifdef CONFIG_SOFTWARE_SUSPEND
533	case LINUX_REBOOT_CMD_SW_SUSPEND:	533	case LINUX_REBOOT_CMD_SW_SUSPEND:
534	{	534	{
535	int ret = software_suspend();	535	int ret = software_suspend();
536	unlock_kernel();	536	unlock_kernel();
537	return ret;	537	return ret;
538	}	538	}
539	#endif	539	#endif
540		540
541	default:	541	default:
542	unlock_kernel();	542	unlock_kernel();
543	return -EINVAL;	543	return -EINVAL;
544	}	544	}
545	unlock_kernel();	545	unlock_kernel();
546	return 0;	546	return 0;
547	}	547	}
548		548
549	static void deferred_cad(void *dummy)	549	static void deferred_cad(void *dummy)
550	{	550	{
551	kernel_restart(NULL);	551	kernel_restart(NULL);
552	}	552	}
553		553
554	/*	554	/*
555	* This function gets called by ctrl-alt-del - ie the keyboard interrupt.	555	* This function gets called by ctrl-alt-del - ie the keyboard interrupt.
556	* As it's called within an interrupt, it may NOT sync: the only choice	556	* As it's called within an interrupt, it may NOT sync: the only choice
557	* is whether to reboot at once, or just ignore the ctrl-alt-del.	557	* is whether to reboot at once, or just ignore the ctrl-alt-del.
558	*/	558	*/
559	void ctrl_alt_del(void)	559	void ctrl_alt_del(void)
560	{	560	{
561	static DECLARE_WORK(cad_work, deferred_cad, NULL);	561	static DECLARE_WORK(cad_work, deferred_cad, NULL);
562		562
563	if (C_A_D)	563	if (C_A_D)
564	schedule_work(&cad_work);	564	schedule_work(&cad_work);
565	else	565	else
566	kill_proc(cad_pid, SIGINT, 1);	566	kill_proc(cad_pid, SIGINT, 1);
567	}	567	}
568		568
569		569
570	/*	570	/*
571	* Unprivileged users may change the real gid to the effective gid	571	* Unprivileged users may change the real gid to the effective gid
572	* or vice versa. (BSD-style)	572	* or vice versa. (BSD-style)
573	*	573	*
574	* If you set the real gid at all, or set the effective gid to a value not	574	* If you set the real gid at all, or set the effective gid to a value not
575	* equal to the real gid, then the saved gid is set to the new effective gid.	575	* equal to the real gid, then the saved gid is set to the new effective gid.
576	*	576	*
577	* This makes it possible for a setgid program to completely drop its	577	* This makes it possible for a setgid program to completely drop its
578	* privileges, which is often a useful assertion to make when you are doing	578	* privileges, which is often a useful assertion to make when you are doing
579	* a security audit over a program.	579	* a security audit over a program.
580	*	580	*
581	* The general idea is that a program which uses just setregid() will be	581	* The general idea is that a program which uses just setregid() will be
582	* 100% compatible with BSD. A program which uses just setgid() will be	582	* 100% compatible with BSD. A program which uses just setgid() will be
583	* 100% compatible with POSIX with saved IDs.	583	* 100% compatible with POSIX with saved IDs.
584	*	584	*
585	* SMP: There are not races, the GIDs are checked only by filesystem	585	* SMP: There are not races, the GIDs are checked only by filesystem
586	* operations (as far as semantic preservation is concerned).	586	* operations (as far as semantic preservation is concerned).
587	*/	587	*/
588	asmlinkage long sys_setregid(gid_t rgid, gid_t egid)	588	asmlinkage long sys_setregid(gid_t rgid, gid_t egid)
589	{	589	{
590	int old_rgid = current->gid;	590	int old_rgid = current->gid;
591	int old_egid = current->egid;	591	int old_egid = current->egid;
592	int new_rgid = old_rgid;	592	int new_rgid = old_rgid;
593	int new_egid = old_egid;	593	int new_egid = old_egid;
594	int retval;	594	int retval;
595		595
596	retval = security_task_setgid(rgid, egid, (gid_t)-1, LSM_SETID_RE);	596	retval = security_task_setgid(rgid, egid, (gid_t)-1, LSM_SETID_RE);
597	if (retval)	597	if (retval)
598	return retval;	598	return retval;
599		599
600	if (rgid != (gid_t) -1) {	600	if (rgid != (gid_t) -1) {
601	if ((old_rgid == rgid) \|\|	601	if ((old_rgid == rgid) \|\|
602	(current->egid==rgid) \|\|	602	(current->egid==rgid) \|\|
603	capable(CAP_SETGID))	603	capable(CAP_SETGID))
604	new_rgid = rgid;	604	new_rgid = rgid;
605	else	605	else
606	return -EPERM;	606	return -EPERM;
607	}	607	}
608	if (egid != (gid_t) -1) {	608	if (egid != (gid_t) -1) {
609	if ((old_rgid == egid) \|\|	609	if ((old_rgid == egid) \|\|
610	(current->egid == egid) \|\|	610	(current->egid == egid) \|\|
611	(current->sgid == egid) \|\|	611	(current->sgid == egid) \|\|
612	capable(CAP_SETGID))	612	capable(CAP_SETGID))
613	new_egid = egid;	613	new_egid = egid;
614	else {	614	else {
615	return -EPERM;	615	return -EPERM;
616	}	616	}
617	}	617	}
618	if (new_egid != old_egid)	618	if (new_egid != old_egid)
619	{	619	{
620	current->mm->dumpable = suid_dumpable;	620	current->mm->dumpable = suid_dumpable;
621	smp_wmb();	621	smp_wmb();
622	}	622	}
623	if (rgid != (gid_t) -1 \|\|	623	if (rgid != (gid_t) -1 \|\|
624	(egid != (gid_t) -1 && egid != old_rgid))	624	(egid != (gid_t) -1 && egid != old_rgid))
625	current->sgid = new_egid;	625	current->sgid = new_egid;
626	current->fsgid = new_egid;	626	current->fsgid = new_egid;
627	current->egid = new_egid;	627	current->egid = new_egid;
628	current->gid = new_rgid;	628	current->gid = new_rgid;
629	key_fsgid_changed(current);	629	key_fsgid_changed(current);
630	proc_id_connector(current, PROC_EVENT_GID);	630	proc_id_connector(current, PROC_EVENT_GID);
631	return 0;	631	return 0;
632	}	632	}
633		633
634	/*	634	/*
635	* setgid() is implemented like SysV w/ SAVED_IDS	635	* setgid() is implemented like SysV w/ SAVED_IDS
636	*	636	*
637	* SMP: Same implicit races as above.	637	* SMP: Same implicit races as above.
638	*/	638	*/
639	asmlinkage long sys_setgid(gid_t gid)	639	asmlinkage long sys_setgid(gid_t gid)
640	{	640	{
641	int old_egid = current->egid;	641	int old_egid = current->egid;
642	int retval;	642	int retval;
643		643
644	retval = security_task_setgid(gid, (gid_t)-1, (gid_t)-1, LSM_SETID_ID);	644	retval = security_task_setgid(gid, (gid_t)-1, (gid_t)-1, LSM_SETID_ID);
645	if (retval)	645	if (retval)
646	return retval;	646	return retval;
647		647
648	if (capable(CAP_SETGID))	648	if (capable(CAP_SETGID))
649	{	649	{
650	if(old_egid != gid)	650	if(old_egid != gid)
651	{	651	{
652	current->mm->dumpable = suid_dumpable;	652	current->mm->dumpable = suid_dumpable;
653	smp_wmb();	653	smp_wmb();
654	}	654	}
655	current->gid = current->egid = current->sgid = current->fsgid = gid;	655	current->gid = current->egid = current->sgid = current->fsgid = gid;
656	}	656	}
657	else if ((gid == current->gid) \|\| (gid == current->sgid))	657	else if ((gid == current->gid) \|\| (gid == current->sgid))
658	{	658	{
659	if(old_egid != gid)	659	if(old_egid != gid)
660	{	660	{
661	current->mm->dumpable = suid_dumpable;	661	current->mm->dumpable = suid_dumpable;
662	smp_wmb();	662	smp_wmb();
663	}	663	}
664	current->egid = current->fsgid = gid;	664	current->egid = current->fsgid = gid;
665	}	665	}
666	else	666	else
667	return -EPERM;	667	return -EPERM;
668		668
669	key_fsgid_changed(current);	669	key_fsgid_changed(current);
670	proc_id_connector(current, PROC_EVENT_GID);	670	proc_id_connector(current, PROC_EVENT_GID);
671	return 0;	671	return 0;
672	}	672	}
673		673
674	static int set_user(uid_t new_ruid, int dumpclear)	674	static int set_user(uid_t new_ruid, int dumpclear)
675	{	675	{
676	struct user_struct *new_user;	676	struct user_struct *new_user;
677		677
678	new_user = alloc_uid(new_ruid);	678	new_user = alloc_uid(new_ruid);
679	if (!new_user)	679	if (!new_user)
680	return -EAGAIN;	680	return -EAGAIN;
681		681
682	if (atomic_read(&new_user->processes) >=	682	if (atomic_read(&new_user->processes) >=
683	current->signal->rlim[RLIMIT_NPROC].rlim_cur &&	683	current->signal->rlim[RLIMIT_NPROC].rlim_cur &&
684	new_user != &root_user) {	684	new_user != &root_user) {
685	free_uid(new_user);	685	free_uid(new_user);
686	return -EAGAIN;	686	return -EAGAIN;
687	}	687	}
688		688
689	switch_uid(new_user);	689	switch_uid(new_user);
690		690
691	if(dumpclear)	691	if(dumpclear)
692	{	692	{
693	current->mm->dumpable = suid_dumpable;	693	current->mm->dumpable = suid_dumpable;
694	smp_wmb();	694	smp_wmb();
695	}	695	}
696	current->uid = new_ruid;	696	current->uid = new_ruid;
697	return 0;	697	return 0;
698	}	698	}
699		699
700	/*	700	/*
701	* Unprivileged users may change the real uid to the effective uid	701	* Unprivileged users may change the real uid to the effective uid
702	* or vice versa. (BSD-style)	702	* or vice versa. (BSD-style)
703	*	703	*
704	* If you set the real uid at all, or set the effective uid to a value not	704	* If you set the real uid at all, or set the effective uid to a value not
705	* equal to the real uid, then the saved uid is set to the new effective uid.	705	* equal to the real uid, then the saved uid is set to the new effective uid.
706	*	706	*
707	* This makes it possible for a setuid program to completely drop its	707	* This makes it possible for a setuid program to completely drop its
708	* privileges, which is often a useful assertion to make when you are doing	708	* privileges, which is often a useful assertion to make when you are doing
709	* a security audit over a program.	709	* a security audit over a program.
710	*	710	*
711	* The general idea is that a program which uses just setreuid() will be	711	* The general idea is that a program which uses just setreuid() will be
712	* 100% compatible with BSD. A program which uses just setuid() will be	712	* 100% compatible with BSD. A program which uses just setuid() will be
713	* 100% compatible with POSIX with saved IDs.	713	* 100% compatible with POSIX with saved IDs.
714	*/	714	*/
715	asmlinkage long sys_setreuid(uid_t ruid, uid_t euid)	715	asmlinkage long sys_setreuid(uid_t ruid, uid_t euid)
716	{	716	{
717	int old_ruid, old_euid, old_suid, new_ruid, new_euid;	717	int old_ruid, old_euid, old_suid, new_ruid, new_euid;
718	int retval;	718	int retval;
719		719
720	retval = security_task_setuid(ruid, euid, (uid_t)-1, LSM_SETID_RE);	720	retval = security_task_setuid(ruid, euid, (uid_t)-1, LSM_SETID_RE);
721	if (retval)	721	if (retval)
722	return retval;	722	return retval;
723		723
724	new_ruid = old_ruid = current->uid;	724	new_ruid = old_ruid = current->uid;
725	new_euid = old_euid = current->euid;	725	new_euid = old_euid = current->euid;
726	old_suid = current->suid;	726	old_suid = current->suid;
727		727
728	if (ruid != (uid_t) -1) {	728	if (ruid != (uid_t) -1) {
729	new_ruid = ruid;	729	new_ruid = ruid;
730	if ((old_ruid != ruid) &&	730	if ((old_ruid != ruid) &&
731	(current->euid != ruid) &&	731	(current->euid != ruid) &&
732	!capable(CAP_SETUID))	732	!capable(CAP_SETUID))
733	return -EPERM;	733	return -EPERM;
734	}	734	}
735		735
736	if (euid != (uid_t) -1) {	736	if (euid != (uid_t) -1) {
737	new_euid = euid;	737	new_euid = euid;
738	if ((old_ruid != euid) &&	738	if ((old_ruid != euid) &&
739	(current->euid != euid) &&	739	(current->euid != euid) &&
740	(current->suid != euid) &&	740	(current->suid != euid) &&
741	!capable(CAP_SETUID))	741	!capable(CAP_SETUID))
742	return -EPERM;	742	return -EPERM;
743	}	743	}
744		744
745	if (new_ruid != old_ruid && set_user(new_ruid, new_euid != old_euid) < 0)	745	if (new_ruid != old_ruid && set_user(new_ruid, new_euid != old_euid) < 0)
746	return -EAGAIN;	746	return -EAGAIN;
747		747
748	if (new_euid != old_euid)	748	if (new_euid != old_euid)
749	{	749	{
750	current->mm->dumpable = suid_dumpable;	750	current->mm->dumpable = suid_dumpable;
751	smp_wmb();	751	smp_wmb();
752	}	752	}
753	current->fsuid = current->euid = new_euid;	753	current->fsuid = current->euid = new_euid;
754	if (ruid != (uid_t) -1 \|\|	754	if (ruid != (uid_t) -1 \|\|
755	(euid != (uid_t) -1 && euid != old_ruid))	755	(euid != (uid_t) -1 && euid != old_ruid))
756	current->suid = current->euid;	756	current->suid = current->euid;
757	current->fsuid = current->euid;	757	current->fsuid = current->euid;
758		758
759	key_fsuid_changed(current);	759	key_fsuid_changed(current);
760	proc_id_connector(current, PROC_EVENT_UID);	760	proc_id_connector(current, PROC_EVENT_UID);
761		761
762	return security_task_post_setuid(old_ruid, old_euid, old_suid, LSM_SETID_RE);	762	return security_task_post_setuid(old_ruid, old_euid, old_suid, LSM_SETID_RE);
763	}	763	}
764		764
765		765
766		766
767	/*	767	/*
768	* setuid() is implemented like SysV with SAVED_IDS	768	* setuid() is implemented like SysV with SAVED_IDS
769	*	769	*
770	* Note that SAVED_ID's is deficient in that a setuid root program	770	* Note that SAVED_ID's is deficient in that a setuid root program
771	* like sendmail, for example, cannot set its uid to be a normal	771	* like sendmail, for example, cannot set its uid to be a normal
772	* user and then switch back, because if you're root, setuid() sets	772	* user and then switch back, because if you're root, setuid() sets
773	* the saved uid too. If you don't like this, blame the bright people	773	* the saved uid too. If you don't like this, blame the bright people
774	* in the POSIX committee and/or USG. Note that the BSD-style setreuid()	774	* in the POSIX committee and/or USG. Note that the BSD-style setreuid()
775	* will allow a root program to temporarily drop privileges and be able to	775	* will allow a root program to temporarily drop privileges and be able to
776	* regain them by swapping the real and effective uid.	776	* regain them by swapping the real and effective uid.
777	*/	777	*/
778	asmlinkage long sys_setuid(uid_t uid)	778	asmlinkage long sys_setuid(uid_t uid)
779	{	779	{
780	int old_euid = current->euid;	780	int old_euid = current->euid;
781	int old_ruid, old_suid, new_ruid, new_suid;	781	int old_ruid, old_suid, new_ruid, new_suid;
782	int retval;	782	int retval;
783		783
784	retval = security_task_setuid(uid, (uid_t)-1, (uid_t)-1, LSM_SETID_ID);	784	retval = security_task_setuid(uid, (uid_t)-1, (uid_t)-1, LSM_SETID_ID);
785	if (retval)	785	if (retval)
786	return retval;	786	return retval;
787		787
788	old_ruid = new_ruid = current->uid;	788	old_ruid = new_ruid = current->uid;
789	old_suid = current->suid;	789	old_suid = current->suid;
790	new_suid = old_suid;	790	new_suid = old_suid;
791		791
792	if (capable(CAP_SETUID)) {	792	if (capable(CAP_SETUID)) {
793	if (uid != old_ruid && set_user(uid, old_euid != uid) < 0)	793	if (uid != old_ruid && set_user(uid, old_euid != uid) < 0)
794	return -EAGAIN;	794	return -EAGAIN;
795	new_suid = uid;	795	new_suid = uid;
796	} else if ((uid != current->uid) && (uid != new_suid))	796	} else if ((uid != current->uid) && (uid != new_suid))
797	return -EPERM;	797	return -EPERM;
798		798
799	if (old_euid != uid)	799	if (old_euid != uid)
800	{	800	{
801	current->mm->dumpable = suid_dumpable;	801	current->mm->dumpable = suid_dumpable;
802	smp_wmb();	802	smp_wmb();
803	}	803	}
804	current->fsuid = current->euid = uid;	804	current->fsuid = current->euid = uid;
805	current->suid = new_suid;	805	current->suid = new_suid;
806		806
807	key_fsuid_changed(current);	807	key_fsuid_changed(current);
808	proc_id_connector(current, PROC_EVENT_UID);	808	proc_id_connector(current, PROC_EVENT_UID);
809		809
810	return security_task_post_setuid(old_ruid, old_euid, old_suid, LSM_SETID_ID);	810	return security_task_post_setuid(old_ruid, old_euid, old_suid, LSM_SETID_ID);
811	}	811	}
812		812
813		813
814	/*	814	/*
815	* This function implements a generic ability to update ruid, euid,	815	* This function implements a generic ability to update ruid, euid,
816	* and suid. This allows you to implement the 4.4 compatible seteuid().	816	* and suid. This allows you to implement the 4.4 compatible seteuid().
817	*/	817	*/
818	asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid)	818	asmlinkage long sys_setresuid(uid_t ruid, uid_t euid, uid_t suid)
819	{	819	{
820	int old_ruid = current->uid;	820	int old_ruid = current->uid;
821	int old_euid = current->euid;	821	int old_euid = current->euid;
822	int old_suid = current->suid;	822	int old_suid = current->suid;
823	int retval;	823	int retval;
824		824
825	retval = security_task_setuid(ruid, euid, suid, LSM_SETID_RES);	825	retval = security_task_setuid(ruid, euid, suid, LSM_SETID_RES);
826	if (retval)	826	if (retval)
827	return retval;	827	return retval;
828		828
829	if (!capable(CAP_SETUID)) {	829	if (!capable(CAP_SETUID)) {
830	if ((ruid != (uid_t) -1) && (ruid != current->uid) &&	830	if ((ruid != (uid_t) -1) && (ruid != current->uid) &&
831	(ruid != current->euid) && (ruid != current->suid))	831	(ruid != current->euid) && (ruid != current->suid))
832	return -EPERM;	832	return -EPERM;
833	if ((euid != (uid_t) -1) && (euid != current->uid) &&	833	if ((euid != (uid_t) -1) && (euid != current->uid) &&
834	(euid != current->euid) && (euid != current->suid))	834	(euid != current->euid) && (euid != current->suid))
835	return -EPERM;	835	return -EPERM;
836	if ((suid != (uid_t) -1) && (suid != current->uid) &&	836	if ((suid != (uid_t) -1) && (suid != current->uid) &&
837	(suid != current->euid) && (suid != current->suid))	837	(suid != current->euid) && (suid != current->suid))
838	return -EPERM;	838	return -EPERM;
839	}	839	}
840	if (ruid != (uid_t) -1) {	840	if (ruid != (uid_t) -1) {
841	if (ruid != current->uid && set_user(ruid, euid != current->euid) < 0)	841	if (ruid != current->uid && set_user(ruid, euid != current->euid) < 0)
842	return -EAGAIN;	842	return -EAGAIN;
843	}	843	}
844	if (euid != (uid_t) -1) {	844	if (euid != (uid_t) -1) {
845	if (euid != current->euid)	845	if (euid != current->euid)
846	{	846	{
847	current->mm->dumpable = suid_dumpable;	847	current->mm->dumpable = suid_dumpable;
848	smp_wmb();	848	smp_wmb();
849	}	849	}
850	current->euid = euid;	850	current->euid = euid;
851	}	851	}
852	current->fsuid = current->euid;	852	current->fsuid = current->euid;
853	if (suid != (uid_t) -1)	853	if (suid != (uid_t) -1)
854	current->suid = suid;	854	current->suid = suid;
855		855
856	key_fsuid_changed(current);	856	key_fsuid_changed(current);
857	proc_id_connector(current, PROC_EVENT_UID);	857	proc_id_connector(current, PROC_EVENT_UID);
858		858
859	return security_task_post_setuid(old_ruid, old_euid, old_suid, LSM_SETID_RES);	859	return security_task_post_setuid(old_ruid, old_euid, old_suid, LSM_SETID_RES);
860	}	860	}
861		861
862	asmlinkage long sys_getresuid(uid_t __user ruid, uid_t __user euid, uid_t __user *suid)	862	asmlinkage long sys_getresuid(uid_t __user ruid, uid_t __user euid, uid_t __user *suid)
863	{	863	{
864	int retval;	864	int retval;
865		865
866	if (!(retval = put_user(current->uid, ruid)) &&	866	if (!(retval = put_user(current->uid, ruid)) &&
867	!(retval = put_user(current->euid, euid)))	867	!(retval = put_user(current->euid, euid)))
868	retval = put_user(current->suid, suid);	868	retval = put_user(current->suid, suid);
869		869
870	return retval;	870	return retval;
871	}	871	}
872		872
873	/*	873	/*
874	* Same as above, but for rgid, egid, sgid.	874	* Same as above, but for rgid, egid, sgid.
875	*/	875	*/
876	asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid)	876	asmlinkage long sys_setresgid(gid_t rgid, gid_t egid, gid_t sgid)
877	{	877	{
878	int retval;	878	int retval;
879		879
880	retval = security_task_setgid(rgid, egid, sgid, LSM_SETID_RES);	880	retval = security_task_setgid(rgid, egid, sgid, LSM_SETID_RES);
881	if (retval)	881	if (retval)
882	return retval;	882	return retval;
883		883
884	if (!capable(CAP_SETGID)) {	884	if (!capable(CAP_SETGID)) {
885	if ((rgid != (gid_t) -1) && (rgid != current->gid) &&	885	if ((rgid != (gid_t) -1) && (rgid != current->gid) &&
886	(rgid != current->egid) && (rgid != current->sgid))	886	(rgid != current->egid) && (rgid != current->sgid))
887	return -EPERM;	887	return -EPERM;
888	if ((egid != (gid_t) -1) && (egid != current->gid) &&	888	if ((egid != (gid_t) -1) && (egid != current->gid) &&
889	(egid != current->egid) && (egid != current->sgid))	889	(egid != current->egid) && (egid != current->sgid))
890	return -EPERM;	890	return -EPERM;
891	if ((sgid != (gid_t) -1) && (sgid != current->gid) &&	891	if ((sgid != (gid_t) -1) && (sgid != current->gid) &&
892	(sgid != current->egid) && (sgid != current->sgid))	892	(sgid != current->egid) && (sgid != current->sgid))
893	return -EPERM;	893	return -EPERM;
894	}	894	}
895	if (egid != (gid_t) -1) {	895	if (egid != (gid_t) -1) {
896	if (egid != current->egid)	896	if (egid != current->egid)
897	{	897	{
898	current->mm->dumpable = suid_dumpable;	898	current->mm->dumpable = suid_dumpable;
899	smp_wmb();	899	smp_wmb();
900	}	900	}
901	current->egid = egid;	901	current->egid = egid;
902	}	902	}
903	current->fsgid = current->egid;	903	current->fsgid = current->egid;
904	if (rgid != (gid_t) -1)	904	if (rgid != (gid_t) -1)
905	current->gid = rgid;	905	current->gid = rgid;
906	if (sgid != (gid_t) -1)	906	if (sgid != (gid_t) -1)
907	current->sgid = sgid;	907	current->sgid = sgid;
908		908
909	key_fsgid_changed(current);	909	key_fsgid_changed(current);
910	proc_id_connector(current, PROC_EVENT_GID);	910	proc_id_connector(current, PROC_EVENT_GID);
911	return 0;	911	return 0;
912	}	912	}
913		913
914	asmlinkage long sys_getresgid(gid_t __user rgid, gid_t __user egid, gid_t __user *sgid)	914	asmlinkage long sys_getresgid(gid_t __user rgid, gid_t __user egid, gid_t __user *sgid)
915	{	915	{
916	int retval;	916	int retval;
917		917
918	if (!(retval = put_user(current->gid, rgid)) &&	918	if (!(retval = put_user(current->gid, rgid)) &&
919	!(retval = put_user(current->egid, egid)))	919	!(retval = put_user(current->egid, egid)))
920	retval = put_user(current->sgid, sgid);	920	retval = put_user(current->sgid, sgid);
921		921
922	return retval;	922	return retval;
923	}	923	}
924		924
925		925
926	/*	926	/*
927	* "setfsuid()" sets the fsuid - the uid used for filesystem checks. This	927	* "setfsuid()" sets the fsuid - the uid used for filesystem checks. This
928	* is used for "access()" and for the NFS daemon (letting nfsd stay at	928	* is used for "access()" and for the NFS daemon (letting nfsd stay at
929	* whatever uid it wants to). It normally shadows "euid", except when	929	* whatever uid it wants to). It normally shadows "euid", except when
930	* explicitly set by setfsuid() or for access..	930	* explicitly set by setfsuid() or for access..
931	*/	931	*/
932	asmlinkage long sys_setfsuid(uid_t uid)	932	asmlinkage long sys_setfsuid(uid_t uid)
933	{	933	{
934	int old_fsuid;	934	int old_fsuid;
935		935
936	old_fsuid = current->fsuid;	936	old_fsuid = current->fsuid;
937	if (security_task_setuid(uid, (uid_t)-1, (uid_t)-1, LSM_SETID_FS))	937	if (security_task_setuid(uid, (uid_t)-1, (uid_t)-1, LSM_SETID_FS))
938	return old_fsuid;	938	return old_fsuid;
939		939
940	if (uid == current->uid \|\| uid == current->euid \|\|	940	if (uid == current->uid \|\| uid == current->euid \|\|
941	uid == current->suid \|\| uid == current->fsuid \|\|	941	uid == current->suid \|\| uid == current->fsuid \|\|
942	capable(CAP_SETUID))	942	capable(CAP_SETUID))
943	{	943	{
944	if (uid != old_fsuid)	944	if (uid != old_fsuid)
945	{	945	{
946	current->mm->dumpable = suid_dumpable;	946	current->mm->dumpable = suid_dumpable;
947	smp_wmb();	947	smp_wmb();
948	}	948	}
949	current->fsuid = uid;	949	current->fsuid = uid;
950	}	950	}
951		951
952	key_fsuid_changed(current);	952	key_fsuid_changed(current);
953	proc_id_connector(current, PROC_EVENT_UID);	953	proc_id_connector(current, PROC_EVENT_UID);
954		954
955	security_task_post_setuid(old_fsuid, (uid_t)-1, (uid_t)-1, LSM_SETID_FS);	955	security_task_post_setuid(old_fsuid, (uid_t)-1, (uid_t)-1, LSM_SETID_FS);
956		956
957	return old_fsuid;	957	return old_fsuid;
958	}	958	}
959		959
960	/*	960	/*
961	* Samma på svenska..	961	* Samma på svenska..
962	*/	962	*/
963	asmlinkage long sys_setfsgid(gid_t gid)	963	asmlinkage long sys_setfsgid(gid_t gid)
964	{	964	{
965	int old_fsgid;	965	int old_fsgid;
966		966
967	old_fsgid = current->fsgid;	967	old_fsgid = current->fsgid;
968	if (security_task_setgid(gid, (gid_t)-1, (gid_t)-1, LSM_SETID_FS))	968	if (security_task_setgid(gid, (gid_t)-1, (gid_t)-1, LSM_SETID_FS))
969	return old_fsgid;	969	return old_fsgid;
970		970
971	if (gid == current->gid \|\| gid == current->egid \|\|	971	if (gid == current->gid \|\| gid == current->egid \|\|
972	gid == current->sgid \|\| gid == current->fsgid \|\|	972	gid == current->sgid \|\| gid == current->fsgid \|\|
973	capable(CAP_SETGID))	973	capable(CAP_SETGID))
974	{	974	{
975	if (gid != old_fsgid)	975	if (gid != old_fsgid)
976	{	976	{
977	current->mm->dumpable = suid_dumpable;	977	current->mm->dumpable = suid_dumpable;
978	smp_wmb();	978	smp_wmb();
979	}	979	}
980	current->fsgid = gid;	980	current->fsgid = gid;
981	key_fsgid_changed(current);	981	key_fsgid_changed(current);
982	proc_id_connector(current, PROC_EVENT_GID);	982	proc_id_connector(current, PROC_EVENT_GID);
983	}	983	}
984	return old_fsgid;	984	return old_fsgid;
985	}	985	}
986		986
987	asmlinkage long sys_times(struct tms __user * tbuf)	987	asmlinkage long sys_times(struct tms __user * tbuf)
988	{	988	{
989	/*	989	/*
990	* In the SMP world we might just be unlucky and have one of	990	* In the SMP world we might just be unlucky and have one of
991	* the times increment as we use it. Since the value is an	991	* the times increment as we use it. Since the value is an
992	* atomically safe type this is just fine. Conceptually its	992	* atomically safe type this is just fine. Conceptually its
993	* as if the syscall took an instant longer to occur.	993	* as if the syscall took an instant longer to occur.
994	*/	994	*/
995	if (tbuf) {	995	if (tbuf) {
996	struct tms tmp;	996	struct tms tmp;
997	cputime_t utime, stime, cutime, cstime;	997	cputime_t utime, stime, cutime, cstime;
998		998
999	#ifdef CONFIG_SMP	999	#ifdef CONFIG_SMP
1000	if (thread_group_empty(current)) {	1000	if (thread_group_empty(current)) {
1001	/*	1001	/*
1002	* Single thread case without the use of any locks.	1002	* Single thread case without the use of any locks.
1003	*	1003	*
1004	* We may race with release_task if two threads are	1004	* We may race with release_task if two threads are
1005	* executing. However, release task first adds up the	1005	* executing. However, release task first adds up the
1006	* counters (__exit_signal) before removing the task	1006	* counters (__exit_signal) before removing the task
1007	* from the process tasklist (__unhash_process).	1007	* from the process tasklist (__unhash_process).
1008	* __exit_signal also acquires and releases the	1008	* __exit_signal also acquires and releases the
1009	* siglock which results in the proper memory ordering	1009	* siglock which results in the proper memory ordering
1010	* so that the list modifications are always visible	1010	* so that the list modifications are always visible
1011	* after the counters have been updated.	1011	* after the counters have been updated.
1012	*	1012	*
1013	* If the counters have been updated by the second thread	1013	* If the counters have been updated by the second thread
1014	* but the thread has not yet been removed from the list	1014	* but the thread has not yet been removed from the list
1015	* then the other branch will be executing which will	1015	* then the other branch will be executing which will
1016	* block on tasklist_lock until the exit handling of the	1016	* block on tasklist_lock until the exit handling of the
1017	* other task is finished.	1017	* other task is finished.
1018	*	1018	*
1019	* This also implies that the sighand->siglock cannot	1019	* This also implies that the sighand->siglock cannot
1020	* be held by another processor. So we can also	1020	* be held by another processor. So we can also
1021	* skip acquiring that lock.	1021	* skip acquiring that lock.
1022	*/	1022	*/
1023	utime = cputime_add(current->signal->utime, current->utime);	1023	utime = cputime_add(current->signal->utime, current->utime);
1024	stime = cputime_add(current->signal->utime, current->stime);	1024	stime = cputime_add(current->signal->utime, current->stime);
1025	cutime = current->signal->cutime;	1025	cutime = current->signal->cutime;
1026	cstime = current->signal->cstime;	1026	cstime = current->signal->cstime;
1027	} else	1027	} else
1028	#endif	1028	#endif
1029	{	1029	{
1030		1030
1031	/* Process with multiple threads */	1031	/* Process with multiple threads */
1032	struct task_struct *tsk = current;	1032	struct task_struct *tsk = current;
1033	struct task_struct *t;	1033	struct task_struct *t;
1034		1034
1035	read_lock(&tasklist_lock);	1035	read_lock(&tasklist_lock);
1036	utime = tsk->signal->utime;	1036	utime = tsk->signal->utime;
1037	stime = tsk->signal->stime;	1037	stime = tsk->signal->stime;
1038	t = tsk;	1038	t = tsk;
1039	do {	1039	do {
1040	utime = cputime_add(utime, t->utime);	1040	utime = cputime_add(utime, t->utime);
1041	stime = cputime_add(stime, t->stime);	1041	stime = cputime_add(stime, t->stime);
1042	t = next_thread(t);	1042	t = next_thread(t);
1043	} while (t != tsk);	1043	} while (t != tsk);
1044		1044
1045	/*	1045	/*
1046	* While we have tasklist_lock read-locked, no dying thread	1046	* While we have tasklist_lock read-locked, no dying thread
1047	* can be updating current->signal->[us]time. Instead,	1047	* can be updating current->signal->[us]time. Instead,
1048	* we got their counts included in the live thread loop.	1048	* we got their counts included in the live thread loop.
1049	* However, another thread can come in right now and	1049	* However, another thread can come in right now and
1050	* do a wait call that updates current->signal->c[us]time.	1050	* do a wait call that updates current->signal->c[us]time.
1051	* To make sure we always see that pair updated atomically,	1051	* To make sure we always see that pair updated atomically,
1052	* we take the siglock around fetching them.	1052	* we take the siglock around fetching them.
1053	*/	1053	*/
1054	spin_lock_irq(&tsk->sighand->siglock);	1054	spin_lock_irq(&tsk->sighand->siglock);
1055	cutime = tsk->signal->cutime;	1055	cutime = tsk->signal->cutime;
1056	cstime = tsk->signal->cstime;	1056	cstime = tsk->signal->cstime;
1057	spin_unlock_irq(&tsk->sighand->siglock);	1057	spin_unlock_irq(&tsk->sighand->siglock);
1058	read_unlock(&tasklist_lock);	1058	read_unlock(&tasklist_lock);
1059	}	1059	}
1060	tmp.tms_utime = cputime_to_clock_t(utime);	1060	tmp.tms_utime = cputime_to_clock_t(utime);
1061	tmp.tms_stime = cputime_to_clock_t(stime);	1061	tmp.tms_stime = cputime_to_clock_t(stime);
1062	tmp.tms_cutime = cputime_to_clock_t(cutime);	1062	tmp.tms_cutime = cputime_to_clock_t(cutime);
1063	tmp.tms_cstime = cputime_to_clock_t(cstime);	1063	tmp.tms_cstime = cputime_to_clock_t(cstime);
1064	if (copy_to_user(tbuf, &tmp, sizeof(struct tms)))	1064	if (copy_to_user(tbuf, &tmp, sizeof(struct tms)))
1065	return -EFAULT;	1065	return -EFAULT;
1066	}	1066	}
1067	return (long) jiffies_64_to_clock_t(get_jiffies_64());	1067	return (long) jiffies_64_to_clock_t(get_jiffies_64());
1068	}	1068	}
1069		1069
1070	/*	1070	/*
1071	* This needs some heavy checking ...	1071	* This needs some heavy checking ...
1072	* I just haven't the stomach for it. I also don't fully	1072	* I just haven't the stomach for it. I also don't fully
1073	* understand sessions/pgrp etc. Let somebody who does explain it.	1073	* understand sessions/pgrp etc. Let somebody who does explain it.
1074	*	1074	*
1075	* OK, I think I have the protection semantics right.... this is really	1075	* OK, I think I have the protection semantics right.... this is really
1076	* only important on a multi-user system anyway, to make sure one user	1076	* only important on a multi-user system anyway, to make sure one user
1077	* can't send a signal to a process owned by another. -TYT, 12/12/91	1077	* can't send a signal to a process owned by another. -TYT, 12/12/91
1078	*	1078	*
1079	* Auch. Had to add the 'did_exec' flag to conform completely to POSIX.	1079	* Auch. Had to add the 'did_exec' flag to conform completely to POSIX.
1080	* LBT 04.03.94	1080	* LBT 04.03.94
1081	*/	1081	*/
1082		1082
1083	asmlinkage long sys_setpgid(pid_t pid, pid_t pgid)	1083	asmlinkage long sys_setpgid(pid_t pid, pid_t pgid)
1084	{	1084	{
1085	struct task_struct *p;	1085	struct task_struct *p;
1086	int err = -EINVAL;	1086	int err = -EINVAL;
1087		1087
1088	if (!pid)	1088	if (!pid)
1089	pid = current->pid;	1089	pid = current->pid;
1090	if (!pgid)	1090	if (!pgid)
1091	pgid = pid;	1091	pgid = pid;
1092	if (pgid < 0)	1092	if (pgid < 0)
1093	return -EINVAL;	1093	return -EINVAL;
1094		1094
1095	/* From this point forward we keep holding onto the tasklist lock	1095	/* From this point forward we keep holding onto the tasklist lock
1096	* so that our parent does not change from under us. -DaveM	1096	* so that our parent does not change from under us. -DaveM
1097	*/	1097	*/
1098	write_lock_irq(&tasklist_lock);	1098	write_lock_irq(&tasklist_lock);
1099		1099
1100	err = -ESRCH;	1100	err = -ESRCH;
1101	p = find_task_by_pid(pid);	1101	p = find_task_by_pid(pid);
1102	if (!p)	1102	if (!p)
1103	goto out;	1103	goto out;
1104		1104
1105	err = -EINVAL;	1105	err = -EINVAL;
1106	if (!thread_group_leader(p))	1106	if (!thread_group_leader(p))
1107	goto out;	1107	goto out;
1108		1108
1109	if (p->parent == current \|\| p->real_parent == current) {	1109	if (p->parent == current \|\| p->real_parent == current) {
1110	err = -EPERM;	1110	err = -EPERM;
1111	if (p->signal->session != current->signal->session)	1111	if (p->signal->session != current->signal->session)
1112	goto out;	1112	goto out;
1113	err = -EACCES;	1113	err = -EACCES;
1114	if (p->did_exec)	1114	if (p->did_exec)
1115	goto out;	1115	goto out;
1116	} else {	1116	} else {
1117	err = -ESRCH;	1117	err = -ESRCH;
1118	if (p != current)	1118	if (p != current)
1119	goto out;	1119	goto out;
1120	}	1120	}
1121		1121
1122	err = -EPERM;	1122	err = -EPERM;
1123	if (p->signal->leader)	1123	if (p->signal->leader)
1124	goto out;	1124	goto out;
1125		1125
1126	if (pgid != pid) {	1126	if (pgid != pid) {
1127	struct task_struct *p;	1127	struct task_struct *p;
1128		1128
1129	do_each_task_pid(pgid, PIDTYPE_PGID, p) {	1129	do_each_task_pid(pgid, PIDTYPE_PGID, p) {
1130	if (p->signal->session == current->signal->session)	1130	if (p->signal->session == current->signal->session)
1131	goto ok_pgid;	1131	goto ok_pgid;
1132	} while_each_task_pid(pgid, PIDTYPE_PGID, p);	1132	} while_each_task_pid(pgid, PIDTYPE_PGID, p);
1133	goto out;	1133	goto out;
1134	}	1134	}
1135		1135
1136	ok_pgid:	1136	ok_pgid:
1137	err = security_task_setpgid(p, pgid);	1137	err = security_task_setpgid(p, pgid);
1138	if (err)	1138	if (err)
1139	goto out;	1139	goto out;
1140		1140
1141	if (process_group(p) != pgid) {	1141	if (process_group(p) != pgid) {
1142	detach_pid(p, PIDTYPE_PGID);	1142	detach_pid(p, PIDTYPE_PGID);
1143	p->signal->pgrp = pgid;	1143	p->signal->pgrp = pgid;
1144	attach_pid(p, PIDTYPE_PGID, pgid);	1144	attach_pid(p, PIDTYPE_PGID, pgid);
1145	}	1145	}
1146		1146
1147	err = 0;	1147	err = 0;
1148	out:	1148	out:
1149	/* All paths lead to here, thus we are safe. -DaveM */	1149	/* All paths lead to here, thus we are safe. -DaveM */
1150	write_unlock_irq(&tasklist_lock);	1150	write_unlock_irq(&tasklist_lock);
1151	return err;	1151	return err;
1152	}	1152	}
1153		1153
1154	asmlinkage long sys_getpgid(pid_t pid)	1154	asmlinkage long sys_getpgid(pid_t pid)
1155	{	1155	{
1156	if (!pid) {	1156	if (!pid) {
1157	return process_group(current);	1157	return process_group(current);
1158	} else {	1158	} else {
1159	int retval;	1159	int retval;
1160	struct task_struct *p;	1160	struct task_struct *p;
1161		1161
1162	read_lock(&tasklist_lock);	1162	read_lock(&tasklist_lock);
1163	p = find_task_by_pid(pid);	1163	p = find_task_by_pid(pid);
1164		1164
1165	retval = -ESRCH;	1165	retval = -ESRCH;
1166	if (p) {	1166	if (p) {
1167	retval = security_task_getpgid(p);	1167	retval = security_task_getpgid(p);
1168	if (!retval)	1168	if (!retval)
1169	retval = process_group(p);	1169	retval = process_group(p);
1170	}	1170	}
1171	read_unlock(&tasklist_lock);	1171	read_unlock(&tasklist_lock);
1172	return retval;	1172	return retval;
1173	}	1173	}
1174	}	1174	}
1175		1175
1176	#ifdef __ARCH_WANT_SYS_GETPGRP	1176	#ifdef __ARCH_WANT_SYS_GETPGRP
1177		1177
1178	asmlinkage long sys_getpgrp(void)	1178	asmlinkage long sys_getpgrp(void)
1179	{	1179	{
1180	/* SMP - assuming writes are word atomic this is fine */	1180	/* SMP - assuming writes are word atomic this is fine */
1181	return process_group(current);	1181	return process_group(current);
1182	}	1182	}
1183		1183
1184	#endif	1184	#endif
1185		1185
1186	asmlinkage long sys_getsid(pid_t pid)	1186	asmlinkage long sys_getsid(pid_t pid)
1187	{	1187	{
1188	if (!pid) {	1188	if (!pid) {
1189	return current->signal->session;	1189	return current->signal->session;
1190	} else {	1190	} else {
1191	int retval;	1191	int retval;
1192	struct task_struct *p;	1192	struct task_struct *p;
1193		1193
1194	read_lock(&tasklist_lock);	1194	read_lock(&tasklist_lock);
1195	p = find_task_by_pid(pid);	1195	p = find_task_by_pid(pid);
1196		1196
1197	retval = -ESRCH;	1197	retval = -ESRCH;
1198	if(p) {	1198	if(p) {
1199	retval = security_task_getsid(p);	1199	retval = security_task_getsid(p);
1200	if (!retval)	1200	if (!retval)
1201	retval = p->signal->session;	1201	retval = p->signal->session;
1202	}	1202	}
1203	read_unlock(&tasklist_lock);	1203	read_unlock(&tasklist_lock);
1204	return retval;	1204	return retval;
1205	}	1205	}
1206	}	1206	}
1207		1207
1208	asmlinkage long sys_setsid(void)	1208	asmlinkage long sys_setsid(void)
1209	{	1209	{
1210	struct pid *pid;	1210	struct pid *pid;
1211	int err = -EPERM;	1211	int err = -EPERM;
1212		1212
1213	if (!thread_group_leader(current))	1213	if (!thread_group_leader(current))
1214	return -EINVAL;	1214	return -EINVAL;
1215		1215
1216	down(&tty_sem);	1216	down(&tty_sem);
1217	write_lock_irq(&tasklist_lock);	1217	write_lock_irq(&tasklist_lock);
1218		1218
1219	pid = find_pid(PIDTYPE_PGID, current->pid);	1219	pid = find_pid(PIDTYPE_PGID, current->pid);
1220	if (pid)	1220	if (pid)
1221	goto out;	1221	goto out;
1222		1222
1223	current->signal->leader = 1;	1223	current->signal->leader = 1;
1224	__set_special_pids(current->pid, current->pid);	1224	__set_special_pids(current->pid, current->pid);
1225	current->signal->tty = NULL;	1225	current->signal->tty = NULL;
1226	current->signal->tty_old_pgrp = 0;	1226	current->signal->tty_old_pgrp = 0;
1227	err = process_group(current);	1227	err = process_group(current);
1228	out:	1228	out:
1229	write_unlock_irq(&tasklist_lock);	1229	write_unlock_irq(&tasklist_lock);
1230	up(&tty_sem);	1230	up(&tty_sem);
1231	return err;	1231	return err;
1232	}	1232	}
1233		1233
1234	/*	1234	/*
1235	* Supplementary group IDs	1235	* Supplementary group IDs
1236	*/	1236	*/
1237		1237
1238	/* init to 2 - one for init_task, one to ensure it is never freed */	1238	/* init to 2 - one for init_task, one to ensure it is never freed */
1239	struct group_info init_groups = { .usage = ATOMIC_INIT(2) };	1239	struct group_info init_groups = { .usage = ATOMIC_INIT(2) };
1240		1240
1241	struct group_info *groups_alloc(int gidsetsize)	1241	struct group_info *groups_alloc(int gidsetsize)
1242	{	1242	{
1243	struct group_info *group_info;	1243	struct group_info *group_info;
1244	int nblocks;	1244	int nblocks;
1245	int i;	1245	int i;
1246		1246
1247	nblocks = (gidsetsize + NGROUPS_PER_BLOCK - 1) / NGROUPS_PER_BLOCK;	1247	nblocks = (gidsetsize + NGROUPS_PER_BLOCK - 1) / NGROUPS_PER_BLOCK;
1248	/* Make sure we always allocate at least one indirect block pointer */	1248	/* Make sure we always allocate at least one indirect block pointer */
1249	nblocks = nblocks ? : 1;	1249	nblocks = nblocks ? : 1;
1250	group_info = kmalloc(sizeof(group_info) + nblockssizeof(gid_t *), GFP_USER);	1250	group_info = kmalloc(sizeof(group_info) + nblockssizeof(gid_t *), GFP_USER);
1251	if (!group_info)	1251	if (!group_info)
1252	return NULL;	1252	return NULL;
1253	group_info->ngroups = gidsetsize;	1253	group_info->ngroups = gidsetsize;
1254	group_info->nblocks = nblocks;	1254	group_info->nblocks = nblocks;
1255	atomic_set(&group_info->usage, 1);	1255	atomic_set(&group_info->usage, 1);
1256		1256
1257	if (gidsetsize <= NGROUPS_SMALL) {	1257	if (gidsetsize <= NGROUPS_SMALL) {
1258	group_info->blocks[0] = group_info->small_block;	1258	group_info->blocks[0] = group_info->small_block;
1259	} else {	1259	} else {
1260	for (i = 0; i < nblocks; i++) {	1260	for (i = 0; i < nblocks; i++) {
1261	gid_t *b;	1261	gid_t *b;
1262	b = (void *)__get_free_page(GFP_USER);	1262	b = (void *)__get_free_page(GFP_USER);
1263	if (!b)	1263	if (!b)
1264	goto out_undo_partial_alloc;	1264	goto out_undo_partial_alloc;
1265	group_info->blocks[i] = b;	1265	group_info->blocks[i] = b;
1266	}	1266	}
1267	}	1267	}
1268	return group_info;	1268	return group_info;
1269		1269
1270	out_undo_partial_alloc:	1270	out_undo_partial_alloc:
1271	while (--i >= 0) {	1271	while (--i >= 0) {
1272	free_page((unsigned long)group_info->blocks[i]);	1272	free_page((unsigned long)group_info->blocks[i]);
1273	}	1273	}
1274	kfree(group_info);	1274	kfree(group_info);
1275	return NULL;	1275	return NULL;
1276	}	1276	}
1277		1277
1278	EXPORT_SYMBOL(groups_alloc);	1278	EXPORT_SYMBOL(groups_alloc);
1279		1279
1280	void groups_free(struct group_info *group_info)	1280	void groups_free(struct group_info *group_info)
1281	{	1281	{
1282	if (group_info->blocks[0] != group_info->small_block) {	1282	if (group_info->blocks[0] != group_info->small_block) {
1283	int i;	1283	int i;
1284	for (i = 0; i < group_info->nblocks; i++)	1284	for (i = 0; i < group_info->nblocks; i++)
1285	free_page((unsigned long)group_info->blocks[i]);	1285	free_page((unsigned long)group_info->blocks[i]);
1286	}	1286	}
1287	kfree(group_info);	1287	kfree(group_info);
1288	}	1288	}
1289		1289
1290	EXPORT_SYMBOL(groups_free);	1290	EXPORT_SYMBOL(groups_free);
1291		1291
1292	/* export the group_info to a user-space array */	1292	/* export the group_info to a user-space array */
1293	static int groups_to_user(gid_t __user *grouplist,	1293	static int groups_to_user(gid_t __user *grouplist,
1294	struct group_info *group_info)	1294	struct group_info *group_info)
1295	{	1295	{
1296	int i;	1296	int i;
1297	int count = group_info->ngroups;	1297	int count = group_info->ngroups;
1298		1298
1299	for (i = 0; i < group_info->nblocks; i++) {	1299	for (i = 0; i < group_info->nblocks; i++) {
1300	int cp_count = min(NGROUPS_PER_BLOCK, count);	1300	int cp_count = min(NGROUPS_PER_BLOCK, count);
1301	int off = i * NGROUPS_PER_BLOCK;	1301	int off = i * NGROUPS_PER_BLOCK;
1302	int len = cp_count * sizeof(*grouplist);	1302	int len = cp_count * sizeof(*grouplist);
1303		1303
1304	if (copy_to_user(grouplist+off, group_info->blocks[i], len))	1304	if (copy_to_user(grouplist+off, group_info->blocks[i], len))
1305	return -EFAULT;	1305	return -EFAULT;
1306		1306
1307	count -= cp_count;	1307	count -= cp_count;
1308	}	1308	}
1309	return 0;	1309	return 0;
1310	}	1310	}
1311		1311
1312	/* fill a group_info from a user-space array - it must be allocated already */	1312	/* fill a group_info from a user-space array - it must be allocated already */
1313	static int groups_from_user(struct group_info *group_info,	1313	static int groups_from_user(struct group_info *group_info,
1314	gid_t __user *grouplist)	1314	gid_t __user *grouplist)
1315	{	1315	{
1316	int i;	1316	int i;
1317	int count = group_info->ngroups;	1317	int count = group_info->ngroups;
1318		1318
1319	for (i = 0; i < group_info->nblocks; i++) {	1319	for (i = 0; i < group_info->nblocks; i++) {
1320	int cp_count = min(NGROUPS_PER_BLOCK, count);	1320	int cp_count = min(NGROUPS_PER_BLOCK, count);
1321	int off = i * NGROUPS_PER_BLOCK;	1321	int off = i * NGROUPS_PER_BLOCK;
1322	int len = cp_count * sizeof(*grouplist);	1322	int len = cp_count * sizeof(*grouplist);
1323		1323
1324	if (copy_from_user(group_info->blocks[i], grouplist+off, len))	1324	if (copy_from_user(group_info->blocks[i], grouplist+off, len))
1325	return -EFAULT;	1325	return -EFAULT;
1326		1326
1327	count -= cp_count;	1327	count -= cp_count;
1328	}	1328	}
1329	return 0;	1329	return 0;
1330	}	1330	}
1331		1331
1332	/* a simple Shell sort */	1332	/* a simple Shell sort */
1333	static void groups_sort(struct group_info *group_info)	1333	static void groups_sort(struct group_info *group_info)
1334	{	1334	{
1335	int base, max, stride;	1335	int base, max, stride;
1336	int gidsetsize = group_info->ngroups;	1336	int gidsetsize = group_info->ngroups;
1337		1337
1338	for (stride = 1; stride < gidsetsize; stride = 3 * stride + 1)	1338	for (stride = 1; stride < gidsetsize; stride = 3 * stride + 1)
1339	; /* nothing */	1339	; /* nothing */
1340	stride /= 3;	1340	stride /= 3;
1341		1341
1342	while (stride) {	1342	while (stride) {
1343	max = gidsetsize - stride;	1343	max = gidsetsize - stride;
1344	for (base = 0; base < max; base++) {	1344	for (base = 0; base < max; base++) {
1345	int left = base;	1345	int left = base;
1346	int right = left + stride;	1346	int right = left + stride;
1347	gid_t tmp = GROUP_AT(group_info, right);	1347	gid_t tmp = GROUP_AT(group_info, right);
1348		1348
1349	while (left >= 0 && GROUP_AT(group_info, left) > tmp) {	1349	while (left >= 0 && GROUP_AT(group_info, left) > tmp) {
1350	GROUP_AT(group_info, right) =	1350	GROUP_AT(group_info, right) =
1351	GROUP_AT(group_info, left);	1351	GROUP_AT(group_info, left);
1352	right = left;	1352	right = left;
1353	left -= stride;	1353	left -= stride;
1354	}	1354	}
1355	GROUP_AT(group_info, right) = tmp;	1355	GROUP_AT(group_info, right) = tmp;
1356	}	1356	}
1357	stride /= 3;	1357	stride /= 3;
1358	}	1358	}
1359	}	1359	}
1360		1360
1361	/* a simple bsearch */	1361	/* a simple bsearch */
1362	int groups_search(struct group_info *group_info, gid_t grp)	1362	int groups_search(struct group_info *group_info, gid_t grp)
1363	{	1363	{
1364	int left, right;	1364	int left, right;
1365		1365
1366	if (!group_info)	1366	if (!group_info)
1367	return 0;	1367	return 0;
1368		1368
1369	left = 0;	1369	left = 0;
1370	right = group_info->ngroups;	1370	right = group_info->ngroups;
1371	while (left < right) {	1371	while (left < right) {
1372	int mid = (left+right)/2;	1372	int mid = (left+right)/2;
1373	int cmp = grp - GROUP_AT(group_info, mid);	1373	int cmp = grp - GROUP_AT(group_info, mid);
1374	if (cmp > 0)	1374	if (cmp > 0)
1375	left = mid + 1;	1375	left = mid + 1;
1376	else if (cmp < 0)	1376	else if (cmp < 0)
1377	right = mid;	1377	right = mid;
1378	else	1378	else
1379	return 1;	1379	return 1;
1380	}	1380	}
1381	return 0;	1381	return 0;
1382	}	1382	}
1383		1383
1384	/* validate and set current->group_info */	1384	/* validate and set current->group_info */
1385	int set_current_groups(struct group_info *group_info)	1385	int set_current_groups(struct group_info *group_info)
1386	{	1386	{
1387	int retval;	1387	int retval;
1388	struct group_info *old_info;	1388	struct group_info *old_info;
1389		1389
1390	retval = security_task_setgroups(group_info);	1390	retval = security_task_setgroups(group_info);
1391	if (retval)	1391	if (retval)
1392	return retval;	1392	return retval;
1393		1393
1394	groups_sort(group_info);	1394	groups_sort(group_info);
1395	get_group_info(group_info);	1395	get_group_info(group_info);
1396		1396
1397	task_lock(current);	1397	task_lock(current);
1398	old_info = current->group_info;	1398	old_info = current->group_info;
1399	current->group_info = group_info;	1399	current->group_info = group_info;
1400	task_unlock(current);	1400	task_unlock(current);
1401		1401
1402	put_group_info(old_info);	1402	put_group_info(old_info);
1403		1403
1404	return 0;	1404	return 0;
1405	}	1405	}
1406		1406
1407	EXPORT_SYMBOL(set_current_groups);	1407	EXPORT_SYMBOL(set_current_groups);
1408		1408
1409	asmlinkage long sys_getgroups(int gidsetsize, gid_t __user *grouplist)	1409	asmlinkage long sys_getgroups(int gidsetsize, gid_t __user *grouplist)
1410	{	1410	{
1411	int i = 0;	1411	int i = 0;
1412		1412
1413	/*	1413	/*
1414	* SMP: Nobody else can change our grouplist. Thus we are	1414	* SMP: Nobody else can change our grouplist. Thus we are
1415	* safe.	1415	* safe.
1416	*/	1416	*/
1417		1417
1418	if (gidsetsize < 0)	1418	if (gidsetsize < 0)
1419	return -EINVAL;	1419	return -EINVAL;
1420		1420
1421	/* no need to grab task_lock here; it cannot change */	1421	/* no need to grab task_lock here; it cannot change */
1422	get_group_info(current->group_info);	1422	get_group_info(current->group_info);
1423	i = current->group_info->ngroups;	1423	i = current->group_info->ngroups;
1424	if (gidsetsize) {	1424	if (gidsetsize) {
1425	if (i > gidsetsize) {	1425	if (i > gidsetsize) {
1426	i = -EINVAL;	1426	i = -EINVAL;
1427	goto out;	1427	goto out;
1428	}	1428	}
1429	if (groups_to_user(grouplist, current->group_info)) {	1429	if (groups_to_user(grouplist, current->group_info)) {
1430	i = -EFAULT;	1430	i = -EFAULT;
1431	goto out;	1431	goto out;
1432	}	1432	}
1433	}	1433	}
1434	out:	1434	out:
1435	put_group_info(current->group_info);	1435	put_group_info(current->group_info);
1436	return i;	1436	return i;
1437	}	1437	}
1438		1438
1439	/*	1439	/*
1440	* SMP: Our groups are copy-on-write. We can set them safely	1440	* SMP: Our groups are copy-on-write. We can set them safely
1441	* without another task interfering.	1441	* without another task interfering.
1442	*/	1442	*/
1443		1443
1444	asmlinkage long sys_setgroups(int gidsetsize, gid_t __user *grouplist)	1444	asmlinkage long sys_setgroups(int gidsetsize, gid_t __user *grouplist)
1445	{	1445	{
1446	struct group_info *group_info;	1446	struct group_info *group_info;
1447	int retval;	1447	int retval;
1448		1448
1449	if (!capable(CAP_SETGID))	1449	if (!capable(CAP_SETGID))
1450	return -EPERM;	1450	return -EPERM;
1451	if ((unsigned)gidsetsize > NGROUPS_MAX)	1451	if ((unsigned)gidsetsize > NGROUPS_MAX)
1452	return -EINVAL;	1452	return -EINVAL;
1453		1453
1454	group_info = groups_alloc(gidsetsize);	1454	group_info = groups_alloc(gidsetsize);
1455	if (!group_info)	1455	if (!group_info)
1456	return -ENOMEM;	1456	return -ENOMEM;
1457	retval = groups_from_user(group_info, grouplist);	1457	retval = groups_from_user(group_info, grouplist);
1458	if (retval) {	1458	if (retval) {
1459	put_group_info(group_info);	1459	put_group_info(group_info);
1460	return retval;	1460	return retval;
1461	}	1461	}
1462		1462
1463	retval = set_current_groups(group_info);	1463	retval = set_current_groups(group_info);
1464	put_group_info(group_info);	1464	put_group_info(group_info);
1465		1465
1466	return retval;	1466	return retval;
1467	}	1467	}
1468		1468
1469	/*	1469	/*
1470	* Check whether we're fsgid/egid or in the supplemental group..	1470	* Check whether we're fsgid/egid or in the supplemental group..
1471	*/	1471	*/
1472	int in_group_p(gid_t grp)	1472	int in_group_p(gid_t grp)
1473	{	1473	{
1474	int retval = 1;	1474	int retval = 1;
1475	if (grp != current->fsgid) {	1475	if (grp != current->fsgid) {
1476	get_group_info(current->group_info);	1476	get_group_info(current->group_info);
1477	retval = groups_search(current->group_info, grp);	1477	retval = groups_search(current->group_info, grp);
1478	put_group_info(current->group_info);	1478	put_group_info(current->group_info);
1479	}	1479	}
1480	return retval;	1480	return retval;
1481	}	1481	}
1482		1482
1483	EXPORT_SYMBOL(in_group_p);	1483	EXPORT_SYMBOL(in_group_p);
1484		1484
1485	int in_egroup_p(gid_t grp)	1485	int in_egroup_p(gid_t grp)
1486	{	1486	{
1487	int retval = 1;	1487	int retval = 1;
1488	if (grp != current->egid) {	1488	if (grp != current->egid) {
1489	get_group_info(current->group_info);	1489	get_group_info(current->group_info);
1490	retval = groups_search(current->group_info, grp);	1490	retval = groups_search(current->group_info, grp);
1491	put_group_info(current->group_info);	1491	put_group_info(current->group_info);
1492	}	1492	}
1493	return retval;	1493	return retval;
1494	}	1494	}
1495		1495
1496	EXPORT_SYMBOL(in_egroup_p);	1496	EXPORT_SYMBOL(in_egroup_p);
1497		1497
1498	DECLARE_RWSEM(uts_sem);	1498	DECLARE_RWSEM(uts_sem);
1499		1499
1500	EXPORT_SYMBOL(uts_sem);	1500	EXPORT_SYMBOL(uts_sem);
1501		1501
1502	asmlinkage long sys_newuname(struct new_utsname __user * name)	1502	asmlinkage long sys_newuname(struct new_utsname __user * name)
1503	{	1503	{
1504	int errno = 0;	1504	int errno = 0;
1505		1505
1506	down_read(&uts_sem);	1506	down_read(&uts_sem);
1507	if (copy_to_user(name,&system_utsname,sizeof *name))	1507	if (copy_to_user(name,&system_utsname,sizeof *name))
1508	errno = -EFAULT;	1508	errno = -EFAULT;
1509	up_read(&uts_sem);	1509	up_read(&uts_sem);
1510	return errno;	1510	return errno;
1511	}	1511	}
1512		1512
1513	asmlinkage long sys_sethostname(char __user *name, int len)	1513	asmlinkage long sys_sethostname(char __user *name, int len)
1514	{	1514	{
1515	int errno;	1515	int errno;
1516	char tmp[__NEW_UTS_LEN];	1516	char tmp[__NEW_UTS_LEN];
1517		1517
1518	if (!capable(CAP_SYS_ADMIN))	1518	if (!capable(CAP_SYS_ADMIN))
1519	return -EPERM;	1519	return -EPERM;
1520	if (len < 0 \|\| len > __NEW_UTS_LEN)	1520	if (len < 0 \|\| len > __NEW_UTS_LEN)
1521	return -EINVAL;	1521	return -EINVAL;
1522	down_write(&uts_sem);	1522	down_write(&uts_sem);
1523	errno = -EFAULT;	1523	errno = -EFAULT;
1524	if (!copy_from_user(tmp, name, len)) {	1524	if (!copy_from_user(tmp, name, len)) {
1525	memcpy(system_utsname.nodename, tmp, len);	1525	memcpy(system_utsname.nodename, tmp, len);
1526	system_utsname.nodename[len] = 0;	1526	system_utsname.nodename[len] = 0;
1527	errno = 0;	1527	errno = 0;
1528	}	1528	}
1529	up_write(&uts_sem);	1529	up_write(&uts_sem);
1530	return errno;	1530	return errno;
1531	}	1531	}
1532		1532
1533	#ifdef __ARCH_WANT_SYS_GETHOSTNAME	1533	#ifdef __ARCH_WANT_SYS_GETHOSTNAME
1534		1534
1535	asmlinkage long sys_gethostname(char __user *name, int len)	1535	asmlinkage long sys_gethostname(char __user *name, int len)
1536	{	1536	{
1537	int i, errno;	1537	int i, errno;
1538		1538
1539	if (len < 0)	1539	if (len < 0)
1540	return -EINVAL;	1540	return -EINVAL;
1541	down_read(&uts_sem);	1541	down_read(&uts_sem);
1542	i = 1 + strlen(system_utsname.nodename);	1542	i = 1 + strlen(system_utsname.nodename);
1543	if (i > len)	1543	if (i > len)
1544	i = len;	1544	i = len;
1545	errno = 0;	1545	errno = 0;
1546	if (copy_to_user(name, system_utsname.nodename, i))	1546	if (copy_to_user(name, system_utsname.nodename, i))
1547	errno = -EFAULT;	1547	errno = -EFAULT;
1548	up_read(&uts_sem);	1548	up_read(&uts_sem);
1549	return errno;	1549	return errno;
1550	}	1550	}
1551		1551
1552	#endif	1552	#endif
1553		1553
1554	/*	1554	/*
1555	* Only setdomainname; getdomainname can be implemented by calling	1555	* Only setdomainname; getdomainname can be implemented by calling
1556	* uname()	1556	* uname()
1557	*/	1557	*/
1558	asmlinkage long sys_setdomainname(char __user *name, int len)	1558	asmlinkage long sys_setdomainname(char __user *name, int len)
1559	{	1559	{
1560	int errno;	1560	int errno;
1561	char tmp[__NEW_UTS_LEN];	1561	char tmp[__NEW_UTS_LEN];
1562		1562
1563	if (!capable(CAP_SYS_ADMIN))	1563	if (!capable(CAP_SYS_ADMIN))
1564	return -EPERM;	1564	return -EPERM;
1565	if (len < 0 \|\| len > __NEW_UTS_LEN)	1565	if (len < 0 \|\| len > __NEW_UTS_LEN)
1566	return -EINVAL;	1566	return -EINVAL;
1567		1567
1568	down_write(&uts_sem);	1568	down_write(&uts_sem);
1569	errno = -EFAULT;	1569	errno = -EFAULT;
1570	if (!copy_from_user(tmp, name, len)) {	1570	if (!copy_from_user(tmp, name, len)) {
1571	memcpy(system_utsname.domainname, tmp, len);	1571	memcpy(system_utsname.domainname, tmp, len);
1572	system_utsname.domainname[len] = 0;	1572	system_utsname.domainname[len] = 0;
1573	errno = 0;	1573	errno = 0;
1574	}	1574	}
1575	up_write(&uts_sem);	1575	up_write(&uts_sem);
1576	return errno;	1576	return errno;
1577	}	1577	}
1578		1578
1579	asmlinkage long sys_getrlimit(unsigned int resource, struct rlimit __user *rlim)	1579	asmlinkage long sys_getrlimit(unsigned int resource, struct rlimit __user *rlim)
1580	{	1580	{
1581	if (resource >= RLIM_NLIMITS)	1581	if (resource >= RLIM_NLIMITS)
1582	return -EINVAL;	1582	return -EINVAL;
1583	else {	1583	else {
1584	struct rlimit value;	1584	struct rlimit value;
1585	task_lock(current->group_leader);	1585	task_lock(current->group_leader);
1586	value = current->signal->rlim[resource];	1586	value = current->signal->rlim[resource];
1587	task_unlock(current->group_leader);	1587	task_unlock(current->group_leader);
1588	return copy_to_user(rlim, &value, sizeof(*rlim)) ? -EFAULT : 0;	1588	return copy_to_user(rlim, &value, sizeof(*rlim)) ? -EFAULT : 0;
1589	}	1589	}
1590	}	1590	}
1591		1591
1592	#ifdef __ARCH_WANT_SYS_OLD_GETRLIMIT	1592	#ifdef __ARCH_WANT_SYS_OLD_GETRLIMIT
1593		1593
1594	/*	1594	/*
1595	* Back compatibility for getrlimit. Needed for some apps.	1595	* Back compatibility for getrlimit. Needed for some apps.
1596	*/	1596	*/
1597		1597
1598	asmlinkage long sys_old_getrlimit(unsigned int resource, struct rlimit __user *rlim)	1598	asmlinkage long sys_old_getrlimit(unsigned int resource, struct rlimit __user *rlim)
1599	{	1599	{
1600	struct rlimit x;	1600	struct rlimit x;
1601	if (resource >= RLIM_NLIMITS)	1601	if (resource >= RLIM_NLIMITS)
1602	return -EINVAL;	1602	return -EINVAL;
1603		1603
1604	task_lock(current->group_leader);	1604	task_lock(current->group_leader);
1605	x = current->signal->rlim[resource];	1605	x = current->signal->rlim[resource];
1606	task_unlock(current->group_leader);	1606	task_unlock(current->group_leader);
1607	if(x.rlim_cur > 0x7FFFFFFF)	1607	if(x.rlim_cur > 0x7FFFFFFF)
1608	x.rlim_cur = 0x7FFFFFFF;	1608	x.rlim_cur = 0x7FFFFFFF;
1609	if(x.rlim_max > 0x7FFFFFFF)	1609	if(x.rlim_max > 0x7FFFFFFF)
1610	x.rlim_max = 0x7FFFFFFF;	1610	x.rlim_max = 0x7FFFFFFF;
1611	return copy_to_user(rlim, &x, sizeof(x))?-EFAULT:0;	1611	return copy_to_user(rlim, &x, sizeof(x))?-EFAULT:0;
1612	}	1612	}
1613		1613
1614	#endif	1614	#endif
1615		1615
1616	asmlinkage long sys_setrlimit(unsigned int resource, struct rlimit __user *rlim)	1616	asmlinkage long sys_setrlimit(unsigned int resource, struct rlimit __user *rlim)
1617	{	1617	{
1618	struct rlimit new_rlim, *old_rlim;	1618	struct rlimit new_rlim, *old_rlim;
1619	int retval;	1619	int retval;
1620		1620
1621	if (resource >= RLIM_NLIMITS)	1621	if (resource >= RLIM_NLIMITS)
1622	return -EINVAL;	1622	return -EINVAL;
1623	if(copy_from_user(&new_rlim, rlim, sizeof(*rlim)))	1623	if(copy_from_user(&new_rlim, rlim, sizeof(*rlim)))
1624	return -EFAULT;	1624	return -EFAULT;
1625	if (new_rlim.rlim_cur > new_rlim.rlim_max)	1625	if (new_rlim.rlim_cur > new_rlim.rlim_max)
1626	return -EINVAL;	1626	return -EINVAL;
1627	old_rlim = current->signal->rlim + resource;	1627	old_rlim = current->signal->rlim + resource;
1628	if ((new_rlim.rlim_max > old_rlim->rlim_max) &&	1628	if ((new_rlim.rlim_max > old_rlim->rlim_max) &&
1629	!capable(CAP_SYS_RESOURCE))	1629	!capable(CAP_SYS_RESOURCE))
1630	return -EPERM;	1630	return -EPERM;
1631	if (resource == RLIMIT_NOFILE && new_rlim.rlim_max > NR_OPEN)	1631	if (resource == RLIMIT_NOFILE && new_rlim.rlim_max > NR_OPEN)
1632	return -EPERM;	1632	return -EPERM;
1633		1633
1634	retval = security_task_setrlimit(resource, &new_rlim);	1634	retval = security_task_setrlimit(resource, &new_rlim);
1635	if (retval)	1635	if (retval)
1636	return retval;	1636	return retval;
1637		1637
1638	task_lock(current->group_leader);	1638	task_lock(current->group_leader);
1639	*old_rlim = new_rlim;	1639	*old_rlim = new_rlim;
1640	task_unlock(current->group_leader);	1640	task_unlock(current->group_leader);
1641		1641
1642	if (resource == RLIMIT_CPU && new_rlim.rlim_cur != RLIM_INFINITY &&	1642	if (resource == RLIMIT_CPU && new_rlim.rlim_cur != RLIM_INFINITY &&
1643	(cputime_eq(current->signal->it_prof_expires, cputime_zero) \|\|	1643	(cputime_eq(current->signal->it_prof_expires, cputime_zero) \|\|
1644	new_rlim.rlim_cur <= cputime_to_secs(	1644	new_rlim.rlim_cur <= cputime_to_secs(
1645	current->signal->it_prof_expires))) {	1645	current->signal->it_prof_expires))) {
1646	cputime_t cputime = secs_to_cputime(new_rlim.rlim_cur);	1646	cputime_t cputime = secs_to_cputime(new_rlim.rlim_cur);
1647	read_lock(&tasklist_lock);	1647	read_lock(&tasklist_lock);
1648	spin_lock_irq(&current->sighand->siglock);	1648	spin_lock_irq(&current->sighand->siglock);
1649	set_process_cpu_timer(current, CPUCLOCK_PROF,	1649	set_process_cpu_timer(current, CPUCLOCK_PROF,
1650	&cputime, NULL);	1650	&cputime, NULL);
1651	spin_unlock_irq(&current->sighand->siglock);	1651	spin_unlock_irq(&current->sighand->siglock);
1652	read_unlock(&tasklist_lock);	1652	read_unlock(&tasklist_lock);
1653	}	1653	}
1654		1654
1655	return 0;	1655	return 0;
1656	}	1656	}
1657		1657
1658	/*	1658	/*
1659	* It would make sense to put struct rusage in the task_struct,	1659	* It would make sense to put struct rusage in the task_struct,
1660	* except that would make the task_struct be really big. After	1660	* except that would make the task_struct be really big. After
1661	* task_struct gets moved into malloc'ed memory, it would	1661	* task_struct gets moved into malloc'ed memory, it would
1662	* make sense to do this. It will make moving the rest of the information	1662	* make sense to do this. It will make moving the rest of the information
1663	* a lot simpler! (Which we're not doing right now because we're not	1663	* a lot simpler! (Which we're not doing right now because we're not
1664	* measuring them yet).	1664	* measuring them yet).
1665	*	1665	*
1666	* This expects to be called with tasklist_lock read-locked or better,	1666	* This expects to be called with tasklist_lock read-locked or better,
1667	* and the siglock not locked. It may momentarily take the siglock.	1667	* and the siglock not locked. It may momentarily take the siglock.
1668	*	1668	*
1669	* When sampling multiple threads for RUSAGE_SELF, under SMP we might have	1669	* When sampling multiple threads for RUSAGE_SELF, under SMP we might have
1670	* races with threads incrementing their own counters. But since word	1670	* races with threads incrementing their own counters. But since word
1671	* reads are atomic, we either get new values or old values and we don't	1671	* reads are atomic, we either get new values or old values and we don't
1672	* care which for the sums. We always take the siglock to protect reading	1672	* care which for the sums. We always take the siglock to protect reading
1673	* the c* fields from p->signal from races with exit.c updating those	1673	* the c* fields from p->signal from races with exit.c updating those
1674	* fields when reaping, so a sample either gets all the additions of a	1674	* fields when reaping, so a sample either gets all the additions of a
1675	* given child after it's reaped, or none so this sample is before reaping.	1675	* given child after it's reaped, or none so this sample is before reaping.
1676	*/	1676	*/
1677		1677
1678	static void k_getrusage(struct task_struct p, int who, struct rusage r)	1678	static void k_getrusage(struct task_struct p, int who, struct rusage r)
1679	{	1679	{
1680	struct task_struct *t;	1680	struct task_struct *t;
1681	unsigned long flags;	1681	unsigned long flags;
1682	cputime_t utime, stime;	1682	cputime_t utime, stime;
1683		1683
1684	memset((char ) r, 0, sizeof r);	1684	memset((char ) r, 0, sizeof r);
1685		1685
1686	if (unlikely(!p->signal))	1686	if (unlikely(!p->signal))
1687	return;	1687	return;
1688		1688
1689	switch (who) {	1689	switch (who) {
1690	case RUSAGE_CHILDREN:	1690	case RUSAGE_CHILDREN:
1691	spin_lock_irqsave(&p->sighand->siglock, flags);	1691	spin_lock_irqsave(&p->sighand->siglock, flags);
1692	utime = p->signal->cutime;	1692	utime = p->signal->cutime;
1693	stime = p->signal->cstime;	1693	stime = p->signal->cstime;
1694	r->ru_nvcsw = p->signal->cnvcsw;	1694	r->ru_nvcsw = p->signal->cnvcsw;
1695	r->ru_nivcsw = p->signal->cnivcsw;	1695	r->ru_nivcsw = p->signal->cnivcsw;
1696	r->ru_minflt = p->signal->cmin_flt;	1696	r->ru_minflt = p->signal->cmin_flt;
1697	r->ru_majflt = p->signal->cmaj_flt;	1697	r->ru_majflt = p->signal->cmaj_flt;
1698	spin_unlock_irqrestore(&p->sighand->siglock, flags);	1698	spin_unlock_irqrestore(&p->sighand->siglock, flags);
1699	cputime_to_timeval(utime, &r->ru_utime);	1699	cputime_to_timeval(utime, &r->ru_utime);
1700	cputime_to_timeval(stime, &r->ru_stime);	1700	cputime_to_timeval(stime, &r->ru_stime);
1701	break;	1701	break;
1702	case RUSAGE_SELF:	1702	case RUSAGE_SELF:
1703	spin_lock_irqsave(&p->sighand->siglock, flags);	1703	spin_lock_irqsave(&p->sighand->siglock, flags);
1704	utime = stime = cputime_zero;	1704	utime = stime = cputime_zero;
1705	goto sum_group;	1705	goto sum_group;
1706	case RUSAGE_BOTH:	1706	case RUSAGE_BOTH:
1707	spin_lock_irqsave(&p->sighand->siglock, flags);	1707	spin_lock_irqsave(&p->sighand->siglock, flags);
1708	utime = p->signal->cutime;	1708	utime = p->signal->cutime;
1709	stime = p->signal->cstime;	1709	stime = p->signal->cstime;
1710	r->ru_nvcsw = p->signal->cnvcsw;	1710	r->ru_nvcsw = p->signal->cnvcsw;
1711	r->ru_nivcsw = p->signal->cnivcsw;	1711	r->ru_nivcsw = p->signal->cnivcsw;
1712	r->ru_minflt = p->signal->cmin_flt;	1712	r->ru_minflt = p->signal->cmin_flt;
1713	r->ru_majflt = p->signal->cmaj_flt;	1713	r->ru_majflt = p->signal->cmaj_flt;
1714	sum_group:	1714	sum_group:
1715	utime = cputime_add(utime, p->signal->utime);	1715	utime = cputime_add(utime, p->signal->utime);
1716	stime = cputime_add(stime, p->signal->stime);	1716	stime = cputime_add(stime, p->signal->stime);
1717	r->ru_nvcsw += p->signal->nvcsw;	1717	r->ru_nvcsw += p->signal->nvcsw;
1718	r->ru_nivcsw += p->signal->nivcsw;	1718	r->ru_nivcsw += p->signal->nivcsw;
1719	r->ru_minflt += p->signal->min_flt;	1719	r->ru_minflt += p->signal->min_flt;
1720	r->ru_majflt += p->signal->maj_flt;	1720	r->ru_majflt += p->signal->maj_flt;
1721	t = p;	1721	t = p;
1722	do {	1722	do {
1723	utime = cputime_add(utime, t->utime);	1723	utime = cputime_add(utime, t->utime);
1724	stime = cputime_add(stime, t->stime);	1724	stime = cputime_add(stime, t->stime);
1725	r->ru_nvcsw += t->nvcsw;	1725	r->ru_nvcsw += t->nvcsw;
1726	r->ru_nivcsw += t->nivcsw;	1726	r->ru_nivcsw += t->nivcsw;
1727	r->ru_minflt += t->min_flt;	1727	r->ru_minflt += t->min_flt;
1728	r->ru_majflt += t->maj_flt;	1728	r->ru_majflt += t->maj_flt;
1729	t = next_thread(t);	1729	t = next_thread(t);
1730	} while (t != p);	1730	} while (t != p);
1731	spin_unlock_irqrestore(&p->sighand->siglock, flags);	1731	spin_unlock_irqrestore(&p->sighand->siglock, flags);
1732	cputime_to_timeval(utime, &r->ru_utime);	1732	cputime_to_timeval(utime, &r->ru_utime);
1733	cputime_to_timeval(stime, &r->ru_stime);	1733	cputime_to_timeval(stime, &r->ru_stime);
1734	break;	1734	break;
1735	default:	1735	default:
1736	BUG();	1736	BUG();
1737	}	1737	}
1738	}	1738	}
1739		1739
1740	int getrusage(struct task_struct p, int who, struct rusage __user ru)	1740	int getrusage(struct task_struct p, int who, struct rusage __user ru)
1741	{	1741	{
1742	struct rusage r;	1742	struct rusage r;
1743	read_lock(&tasklist_lock);	1743	read_lock(&tasklist_lock);
1744	k_getrusage(p, who, &r);	1744	k_getrusage(p, who, &r);
1745	read_unlock(&tasklist_lock);	1745	read_unlock(&tasklist_lock);
1746	return copy_to_user(ru, &r, sizeof(r)) ? -EFAULT : 0;	1746	return copy_to_user(ru, &r, sizeof(r)) ? -EFAULT : 0;
1747	}	1747	}
1748		1748
1749	asmlinkage long sys_getrusage(int who, struct rusage __user *ru)	1749	asmlinkage long sys_getrusage(int who, struct rusage __user *ru)
1750	{	1750	{
1751	if (who != RUSAGE_SELF && who != RUSAGE_CHILDREN)	1751	if (who != RUSAGE_SELF && who != RUSAGE_CHILDREN)
1752	return -EINVAL;	1752	return -EINVAL;
1753	return getrusage(current, who, ru);	1753	return getrusage(current, who, ru);
1754	}	1754	}
1755		1755
1756	asmlinkage long sys_umask(int mask)	1756	asmlinkage long sys_umask(int mask)
1757	{	1757	{
1758	mask = xchg(&current->fs->umask, mask & S_IRWXUGO);	1758	mask = xchg(&current->fs->umask, mask & S_IRWXUGO);
1759	return mask;	1759	return mask;
1760	}	1760	}
1761		1761
1762	asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3,	1762	asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3,
1763	unsigned long arg4, unsigned long arg5)	1763	unsigned long arg4, unsigned long arg5)
1764	{	1764	{
1765	long error;	1765	long error;
1766		1766
1767	error = security_task_prctl(option, arg2, arg3, arg4, arg5);	1767	error = security_task_prctl(option, arg2, arg3, arg4, arg5);
1768	if (error)	1768	if (error)
1769	return error;	1769	return error;
1770		1770
1771	switch (option) {	1771	switch (option) {
1772	case PR_SET_PDEATHSIG:	1772	case PR_SET_PDEATHSIG:
1773	if (!valid_signal(arg2)) {	1773	if (!valid_signal(arg2)) {
1774	error = -EINVAL;	1774	error = -EINVAL;
1775	break;	1775	break;
1776	}	1776	}
1777	current->pdeath_signal = arg2;	1777	current->pdeath_signal = arg2;
1778	break;	1778	break;
1779	case PR_GET_PDEATHSIG:	1779	case PR_GET_PDEATHSIG:
1780	error = put_user(current->pdeath_signal, (int __user *)arg2);	1780	error = put_user(current->pdeath_signal, (int __user *)arg2);
1781	break;	1781	break;
1782	case PR_GET_DUMPABLE:	1782	case PR_GET_DUMPABLE:
1783	error = current->mm->dumpable;	1783	error = current->mm->dumpable;
1784	break;	1784	break;
1785	case PR_SET_DUMPABLE:	1785	case PR_SET_DUMPABLE:
1786	if (arg2 < 0 \|\| arg2 > 2) {	1786	if (arg2 < 0 \|\| arg2 > 2) {
1787	error = -EINVAL;	1787	error = -EINVAL;
1788	break;	1788	break;
1789	}	1789	}
1790	current->mm->dumpable = arg2;	1790	current->mm->dumpable = arg2;
1791	break;	1791	break;
1792		1792
1793	case PR_SET_UNALIGN:	1793	case PR_SET_UNALIGN:
1794	error = SET_UNALIGN_CTL(current, arg2);	1794	error = SET_UNALIGN_CTL(current, arg2);
1795	break;	1795	break;
1796	case PR_GET_UNALIGN:	1796	case PR_GET_UNALIGN:
1797	error = GET_UNALIGN_CTL(current, arg2);	1797	error = GET_UNALIGN_CTL(current, arg2);
1798	break;	1798	break;
1799	case PR_SET_FPEMU:	1799	case PR_SET_FPEMU:
1800	error = SET_FPEMU_CTL(current, arg2);	1800	error = SET_FPEMU_CTL(current, arg2);
1801	break;	1801	break;
1802	case PR_GET_FPEMU:	1802	case PR_GET_FPEMU:
1803	error = GET_FPEMU_CTL(current, arg2);	1803	error = GET_FPEMU_CTL(current, arg2);
1804	break;	1804	break;
1805	case PR_SET_FPEXC:	1805	case PR_SET_FPEXC:
1806	error = SET_FPEXC_CTL(current, arg2);	1806	error = SET_FPEXC_CTL(current, arg2);
1807	break;	1807	break;
1808	case PR_GET_FPEXC:	1808	case PR_GET_FPEXC:
1809	error = GET_FPEXC_CTL(current, arg2);	1809	error = GET_FPEXC_CTL(current, arg2);
1810	break;	1810	break;
1811	case PR_GET_TIMING:	1811	case PR_GET_TIMING:
1812	error = PR_TIMING_STATISTICAL;	1812	error = PR_TIMING_STATISTICAL;
1813	break;	1813	break;
1814	case PR_SET_TIMING:	1814	case PR_SET_TIMING:
1815	if (arg2 == PR_TIMING_STATISTICAL)	1815	if (arg2 == PR_TIMING_STATISTICAL)
1816	error = 0;	1816	error = 0;
1817	else	1817	else
1818	error = -EINVAL;	1818	error = -EINVAL;
1819	break;	1819	break;
1820		1820
1821	case PR_GET_KEEPCAPS:	1821	case PR_GET_KEEPCAPS:
1822	if (current->keep_capabilities)	1822	if (current->keep_capabilities)
1823	error = 1;	1823	error = 1;
1824	break;	1824	break;
1825	case PR_SET_KEEPCAPS:	1825	case PR_SET_KEEPCAPS:
1826	if (arg2 != 0 && arg2 != 1) {	1826	if (arg2 != 0 && arg2 != 1) {
1827	error = -EINVAL;	1827	error = -EINVAL;
1828	break;	1828	break;
1829	}	1829	}
1830	current->keep_capabilities = arg2;	1830	current->keep_capabilities = arg2;
1831	break;	1831	break;
1832	case PR_SET_NAME: {	1832	case PR_SET_NAME: {
1833	struct task_struct *me = current;	1833	struct task_struct *me = current;
1834	unsigned char ncomm[sizeof(me->comm)];	1834	unsigned char ncomm[sizeof(me->comm)];
1835		1835
1836	ncomm[sizeof(me->comm)-1] = 0;	1836	ncomm[sizeof(me->comm)-1] = 0;
1837	if (strncpy_from_user(ncomm, (char __user *)arg2,	1837	if (strncpy_from_user(ncomm, (char __user *)arg2,
1838	sizeof(me->comm)-1) < 0)	1838	sizeof(me->comm)-1) < 0)
1839	return -EFAULT;	1839	return -EFAULT;
1840	set_task_comm(me, ncomm);	1840	set_task_comm(me, ncomm);
1841	return 0;	1841	return 0;
1842	}	1842	}
1843	case PR_GET_NAME: {	1843	case PR_GET_NAME: {
1844	struct task_struct *me = current;	1844	struct task_struct *me = current;
1845	unsigned char tcomm[sizeof(me->comm)];	1845	unsigned char tcomm[sizeof(me->comm)];
1846		1846
1847	get_task_comm(tcomm, me);	1847	get_task_comm(tcomm, me);
1848	if (copy_to_user((char __user *)arg2, tcomm, sizeof(tcomm)))	1848	if (copy_to_user((char __user *)arg2, tcomm, sizeof(tcomm)))
1849	return -EFAULT;	1849	return -EFAULT;
1850	return 0;	1850	return 0;
1851	}	1851	}
1852	default:	1852	default:
1853	error = -EINVAL;	1853	error = -EINVAL;
1854	break;	1854	break;
1855	}	1855	}
1856	return error;	1856	return error;
1857	}	1857	}
1858		1858