2008-01-29 20:53:40 +07:00
|
|
|
/*
|
|
|
|
* Functions related to setting various queue properties from drivers
|
|
|
|
*/
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/bio.h>
|
|
|
|
#include <linux/blkdev.h>
|
|
|
|
#include <linux/bootmem.h> /* for max_pfn/max_low_pfn */
|
2009-07-31 22:49:12 +07:00
|
|
|
#include <linux/gcd.h>
|
2008-01-29 20:53:40 +07:00
|
|
|
|
|
|
|
#include "blk.h"
|
|
|
|
|
2008-01-31 19:03:55 +07:00
|
|
|
unsigned long blk_max_low_pfn;
|
2008-01-29 20:53:40 +07:00
|
|
|
EXPORT_SYMBOL(blk_max_low_pfn);
|
2008-01-31 19:03:55 +07:00
|
|
|
|
|
|
|
unsigned long blk_max_pfn;
|
2008-01-29 20:53:40 +07:00
|
|
|
|
|
|
|
/**
|
|
|
|
* 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(struct request_queue *q, prep_rq_fn *pfn)
|
|
|
|
{
|
|
|
|
q->prep_rq_fn = pfn;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_prep_rq);
|
|
|
|
|
2008-08-06 00:01:53 +07:00
|
|
|
/**
|
|
|
|
* blk_queue_set_discard - set a discard_sectors function for queue
|
|
|
|
* @q: queue
|
|
|
|
* @dfn: prepare_discard function
|
|
|
|
*
|
|
|
|
* It's possible for a queue to register a discard callback which is used
|
|
|
|
* to transform a discard request into the appropriate type for the
|
|
|
|
* hardware. If none is registered, then discard requests are failed
|
|
|
|
* with %EOPNOTSUPP.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
void blk_queue_set_discard(struct request_queue *q, prepare_discard_fn *dfn)
|
|
|
|
{
|
|
|
|
q->prepare_discard_fn = dfn;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_set_discard);
|
|
|
|
|
2008-01-29 20:53:40 +07:00
|
|
|
/**
|
|
|
|
* 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(struct request_queue *q, merge_bvec_fn *mbfn)
|
|
|
|
{
|
|
|
|
q->merge_bvec_fn = mbfn;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_merge_bvec);
|
|
|
|
|
|
|
|
void blk_queue_softirq_done(struct request_queue *q, softirq_done_fn *fn)
|
|
|
|
{
|
|
|
|
q->softirq_done_fn = fn;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_softirq_done);
|
|
|
|
|
2008-09-14 19:55:09 +07:00
|
|
|
void blk_queue_rq_timeout(struct request_queue *q, unsigned int timeout)
|
|
|
|
{
|
|
|
|
q->rq_timeout = timeout;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(blk_queue_rq_timeout);
|
|
|
|
|
|
|
|
void blk_queue_rq_timed_out(struct request_queue *q, rq_timed_out_fn *fn)
|
|
|
|
{
|
|
|
|
q->rq_timed_out_fn = fn;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(blk_queue_rq_timed_out);
|
|
|
|
|
2008-10-01 21:12:15 +07:00
|
|
|
void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn)
|
|
|
|
{
|
|
|
|
q->lld_busy_fn = fn;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(blk_queue_lld_busy);
|
|
|
|
|
2009-06-16 13:23:52 +07:00
|
|
|
/**
|
|
|
|
* blk_set_default_limits - reset limits to default values
|
2009-06-19 14:18:32 +07:00
|
|
|
* @lim: the queue_limits structure to reset
|
2009-06-16 13:23:52 +07:00
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* Returns a queue_limit struct to its default state. Can be used by
|
|
|
|
* stacking drivers like DM that stage table swaps and reuse an
|
|
|
|
* existing device queue.
|
|
|
|
*/
|
|
|
|
void blk_set_default_limits(struct queue_limits *lim)
|
|
|
|
{
|
|
|
|
lim->max_phys_segments = MAX_PHYS_SEGMENTS;
|
|
|
|
lim->max_hw_segments = MAX_HW_SEGMENTS;
|
|
|
|
lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
|
|
|
|
lim->max_segment_size = MAX_SEGMENT_SIZE;
|
|
|
|
lim->max_sectors = lim->max_hw_sectors = SAFE_MAX_SECTORS;
|
|
|
|
lim->logical_block_size = lim->physical_block_size = lim->io_min = 512;
|
2009-06-18 14:56:03 +07:00
|
|
|
lim->bounce_pfn = (unsigned long)(BLK_BOUNCE_ANY >> PAGE_SHIFT);
|
2009-06-16 13:23:52 +07:00
|
|
|
lim->alignment_offset = 0;
|
|
|
|
lim->io_opt = 0;
|
|
|
|
lim->misaligned = 0;
|
|
|
|
lim->no_cluster = 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_set_default_limits);
|
|
|
|
|
2008-01-29 20:53:40 +07:00
|
|
|
/**
|
|
|
|
* 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.
|
|
|
|
**/
|
2008-01-31 19:03:55 +07:00
|
|
|
void blk_queue_make_request(struct request_queue *q, make_request_fn *mfn)
|
2008-01-29 20:53:40 +07:00
|
|
|
{
|
|
|
|
/*
|
|
|
|
* set defaults
|
|
|
|
*/
|
|
|
|
q->nr_requests = BLKDEV_MAX_RQ;
|
2008-12-03 18:55:08 +07:00
|
|
|
|
2008-01-29 20:53:40 +07:00
|
|
|
q->make_request_fn = mfn;
|
|
|
|
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;
|
|
|
|
|
|
|
|
q->unplug_timer.function = blk_unplug_timeout;
|
|
|
|
q->unplug_timer.data = (unsigned long)q;
|
|
|
|
|
2009-06-16 13:23:52 +07:00
|
|
|
blk_set_default_limits(&q->limits);
|
|
|
|
|
2009-07-28 14:07:29 +07:00
|
|
|
/*
|
|
|
|
* If the caller didn't supply a lock, fall back to our embedded
|
|
|
|
* per-queue locks
|
|
|
|
*/
|
|
|
|
if (!q->queue_lock)
|
|
|
|
q->queue_lock = &q->__queue_lock;
|
|
|
|
|
2008-01-29 20:53:40 +07:00
|
|
|
/*
|
|
|
|
* by default assume old behaviour and bounce for any highmem page
|
|
|
|
*/
|
|
|
|
blk_queue_bounce_limit(q, BLK_BOUNCE_HIGH);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_make_request);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* blk_queue_bounce_limit - set bounce buffer limit for queue
|
2009-04-15 20:10:25 +07:00
|
|
|
* @q: the request queue for the device
|
|
|
|
* @dma_mask: the maximum address the device can handle
|
2008-01-29 20:53:40 +07:00
|
|
|
*
|
|
|
|
* 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
|
2009-04-15 20:10:25 +07:00
|
|
|
* buffers for doing I/O to pages residing above @dma_mask.
|
2008-01-29 20:53:40 +07:00
|
|
|
**/
|
2009-04-15 20:10:25 +07:00
|
|
|
void blk_queue_bounce_limit(struct request_queue *q, u64 dma_mask)
|
2008-01-29 20:53:40 +07:00
|
|
|
{
|
2009-04-15 20:10:25 +07:00
|
|
|
unsigned long b_pfn = dma_mask >> PAGE_SHIFT;
|
2008-01-29 20:53:40 +07:00
|
|
|
int dma = 0;
|
|
|
|
|
|
|
|
q->bounce_gfp = GFP_NOIO;
|
|
|
|
#if BITS_PER_LONG == 64
|
2009-04-15 20:10:25 +07:00
|
|
|
/*
|
|
|
|
* Assume anything <= 4GB can be handled by IOMMU. Actually
|
|
|
|
* some IOMMUs can handle everything, but I don't know of a
|
|
|
|
* way to test this here.
|
|
|
|
*/
|
|
|
|
if (b_pfn < (min_t(u64, 0xffffffffUL, BLK_BOUNCE_HIGH) >> PAGE_SHIFT))
|
2008-01-29 20:53:40 +07:00
|
|
|
dma = 1;
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.bounce_pfn = max_low_pfn;
|
2008-01-29 20:53:40 +07:00
|
|
|
#else
|
2008-01-31 19:03:55 +07:00
|
|
|
if (b_pfn < blk_max_low_pfn)
|
2008-01-29 20:53:40 +07:00
|
|
|
dma = 1;
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.bounce_pfn = b_pfn;
|
2008-01-29 20:53:40 +07:00
|
|
|
#endif
|
|
|
|
if (dma) {
|
|
|
|
init_emergency_isa_pool();
|
|
|
|
q->bounce_gfp = GFP_NOIO | GFP_DMA;
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.bounce_pfn = b_pfn;
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
}
|
|
|
|
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(struct request_queue *q, unsigned int max_sectors)
|
|
|
|
{
|
|
|
|
if ((max_sectors << 9) < PAGE_CACHE_SIZE) {
|
|
|
|
max_sectors = 1 << (PAGE_CACHE_SHIFT - 9);
|
2008-05-01 18:35:17 +07:00
|
|
|
printk(KERN_INFO "%s: set to minimum %d\n",
|
|
|
|
__func__, max_sectors);
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
if (BLK_DEF_MAX_SECTORS > max_sectors)
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.max_hw_sectors = q->limits.max_sectors = max_sectors;
|
2008-01-29 20:53:40 +07:00
|
|
|
else {
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.max_sectors = BLK_DEF_MAX_SECTORS;
|
|
|
|
q->limits.max_hw_sectors = max_sectors;
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_max_sectors);
|
|
|
|
|
2009-05-23 04:17:50 +07:00
|
|
|
void blk_queue_max_hw_sectors(struct request_queue *q, unsigned int max_sectors)
|
|
|
|
{
|
|
|
|
if (BLK_DEF_MAX_SECTORS > max_sectors)
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.max_hw_sectors = BLK_DEF_MAX_SECTORS;
|
2009-05-23 04:17:50 +07:00
|
|
|
else
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.max_hw_sectors = max_sectors;
|
2009-05-23 04:17:50 +07:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_max_hw_sectors);
|
|
|
|
|
2008-01-29 20:53:40 +07:00
|
|
|
/**
|
|
|
|
* 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(struct request_queue *q,
|
|
|
|
unsigned short max_segments)
|
|
|
|
{
|
|
|
|
if (!max_segments) {
|
|
|
|
max_segments = 1;
|
2008-05-01 18:35:17 +07:00
|
|
|
printk(KERN_INFO "%s: set to minimum %d\n",
|
|
|
|
__func__, max_segments);
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.max_phys_segments = max_segments;
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
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
|
2008-08-20 01:13:11 +07:00
|
|
|
* address/length pairs the host adapter can actually give at once
|
2008-01-29 20:53:40 +07:00
|
|
|
* to the device.
|
|
|
|
**/
|
|
|
|
void blk_queue_max_hw_segments(struct request_queue *q,
|
|
|
|
unsigned short max_segments)
|
|
|
|
{
|
|
|
|
if (!max_segments) {
|
|
|
|
max_segments = 1;
|
2008-05-01 18:35:17 +07:00
|
|
|
printk(KERN_INFO "%s: set to minimum %d\n",
|
|
|
|
__func__, max_segments);
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.max_hw_segments = max_segments;
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
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(struct request_queue *q, unsigned int max_size)
|
|
|
|
{
|
|
|
|
if (max_size < PAGE_CACHE_SIZE) {
|
|
|
|
max_size = PAGE_CACHE_SIZE;
|
2008-05-01 18:35:17 +07:00
|
|
|
printk(KERN_INFO "%s: set to minimum %d\n",
|
|
|
|
__func__, max_size);
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.max_segment_size = max_size;
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_max_segment_size);
|
|
|
|
|
|
|
|
/**
|
2009-05-23 04:17:49 +07:00
|
|
|
* blk_queue_logical_block_size - set logical block size for the queue
|
2008-01-29 20:53:40 +07:00
|
|
|
* @q: the request queue for the device
|
2009-05-23 04:17:49 +07:00
|
|
|
* @size: the logical block size, in bytes
|
2008-01-29 20:53:40 +07:00
|
|
|
*
|
|
|
|
* Description:
|
2009-05-23 04:17:49 +07:00
|
|
|
* This should be set to the lowest possible block size that the
|
|
|
|
* storage device can address. The default of 512 covers most
|
|
|
|
* hardware.
|
2008-01-29 20:53:40 +07:00
|
|
|
**/
|
2009-05-23 04:17:49 +07:00
|
|
|
void blk_queue_logical_block_size(struct request_queue *q, unsigned short size)
|
2008-01-29 20:53:40 +07:00
|
|
|
{
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.logical_block_size = size;
|
2009-05-23 04:17:53 +07:00
|
|
|
|
|
|
|
if (q->limits.physical_block_size < size)
|
|
|
|
q->limits.physical_block_size = size;
|
|
|
|
|
|
|
|
if (q->limits.io_min < q->limits.physical_block_size)
|
|
|
|
q->limits.io_min = q->limits.physical_block_size;
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
2009-05-23 04:17:49 +07:00
|
|
|
EXPORT_SYMBOL(blk_queue_logical_block_size);
|
2008-01-29 20:53:40 +07:00
|
|
|
|
2009-05-23 04:17:53 +07:00
|
|
|
/**
|
|
|
|
* blk_queue_physical_block_size - set physical block size for the queue
|
|
|
|
* @q: the request queue for the device
|
|
|
|
* @size: the physical block size, in bytes
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* This should be set to the lowest possible sector size that the
|
|
|
|
* hardware can operate on without reverting to read-modify-write
|
|
|
|
* operations.
|
|
|
|
*/
|
|
|
|
void blk_queue_physical_block_size(struct request_queue *q, unsigned short size)
|
|
|
|
{
|
|
|
|
q->limits.physical_block_size = size;
|
|
|
|
|
|
|
|
if (q->limits.physical_block_size < q->limits.logical_block_size)
|
|
|
|
q->limits.physical_block_size = q->limits.logical_block_size;
|
|
|
|
|
|
|
|
if (q->limits.io_min < q->limits.physical_block_size)
|
|
|
|
q->limits.io_min = q->limits.physical_block_size;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_physical_block_size);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* blk_queue_alignment_offset - set physical block alignment offset
|
|
|
|
* @q: the request queue for the device
|
2009-06-12 10:00:41 +07:00
|
|
|
* @offset: alignment offset in bytes
|
2009-05-23 04:17:53 +07:00
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* Some devices are naturally misaligned to compensate for things like
|
|
|
|
* the legacy DOS partition table 63-sector offset. Low-level drivers
|
|
|
|
* should call this function for devices whose first sector is not
|
|
|
|
* naturally aligned.
|
|
|
|
*/
|
|
|
|
void blk_queue_alignment_offset(struct request_queue *q, unsigned int offset)
|
|
|
|
{
|
|
|
|
q->limits.alignment_offset =
|
|
|
|
offset & (q->limits.physical_block_size - 1);
|
|
|
|
q->limits.misaligned = 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_alignment_offset);
|
|
|
|
|
2009-07-31 22:49:11 +07:00
|
|
|
/**
|
|
|
|
* blk_limits_io_min - set minimum request size for a device
|
|
|
|
* @limits: the queue limits
|
|
|
|
* @min: smallest I/O size in bytes
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* Some devices have an internal block size bigger than the reported
|
|
|
|
* hardware sector size. This function can be used to signal the
|
|
|
|
* smallest I/O the device can perform without incurring a performance
|
|
|
|
* penalty.
|
|
|
|
*/
|
|
|
|
void blk_limits_io_min(struct queue_limits *limits, unsigned int min)
|
|
|
|
{
|
|
|
|
limits->io_min = min;
|
|
|
|
|
|
|
|
if (limits->io_min < limits->logical_block_size)
|
|
|
|
limits->io_min = limits->logical_block_size;
|
|
|
|
|
|
|
|
if (limits->io_min < limits->physical_block_size)
|
|
|
|
limits->io_min = limits->physical_block_size;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_limits_io_min);
|
|
|
|
|
2009-05-23 04:17:53 +07:00
|
|
|
/**
|
|
|
|
* blk_queue_io_min - set minimum request size for the queue
|
|
|
|
* @q: the request queue for the device
|
2009-06-12 10:00:41 +07:00
|
|
|
* @min: smallest I/O size in bytes
|
2009-05-23 04:17:53 +07:00
|
|
|
*
|
|
|
|
* Description:
|
2009-07-31 22:49:13 +07:00
|
|
|
* Storage devices may report a granularity or preferred minimum I/O
|
|
|
|
* size which is the smallest request the device can perform without
|
|
|
|
* incurring a performance penalty. For disk drives this is often the
|
|
|
|
* physical block size. For RAID arrays it is often the stripe chunk
|
|
|
|
* size. A properly aligned multiple of minimum_io_size is the
|
|
|
|
* preferred request size for workloads where a high number of I/O
|
|
|
|
* operations is desired.
|
2009-05-23 04:17:53 +07:00
|
|
|
*/
|
|
|
|
void blk_queue_io_min(struct request_queue *q, unsigned int min)
|
|
|
|
{
|
2009-07-31 22:49:11 +07:00
|
|
|
blk_limits_io_min(&q->limits, min);
|
2009-05-23 04:17:53 +07:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_io_min);
|
|
|
|
|
2009-09-12 02:54:52 +07:00
|
|
|
/**
|
|
|
|
* blk_limits_io_opt - set optimal request size for a device
|
|
|
|
* @limits: the queue limits
|
|
|
|
* @opt: smallest I/O size in bytes
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* Storage devices may report an optimal I/O size, which is the
|
|
|
|
* device's preferred unit for sustained I/O. This is rarely reported
|
|
|
|
* for disk drives. For RAID arrays it is usually the stripe width or
|
|
|
|
* the internal track size. A properly aligned multiple of
|
|
|
|
* optimal_io_size is the preferred request size for workloads where
|
|
|
|
* sustained throughput is desired.
|
|
|
|
*/
|
|
|
|
void blk_limits_io_opt(struct queue_limits *limits, unsigned int opt)
|
|
|
|
{
|
|
|
|
limits->io_opt = opt;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_limits_io_opt);
|
|
|
|
|
2009-05-23 04:17:53 +07:00
|
|
|
/**
|
|
|
|
* blk_queue_io_opt - set optimal request size for the queue
|
|
|
|
* @q: the request queue for the device
|
2009-06-12 10:00:41 +07:00
|
|
|
* @opt: optimal request size in bytes
|
2009-05-23 04:17:53 +07:00
|
|
|
*
|
|
|
|
* Description:
|
2009-07-31 22:49:13 +07:00
|
|
|
* Storage devices may report an optimal I/O size, which is the
|
|
|
|
* device's preferred unit for sustained I/O. This is rarely reported
|
|
|
|
* for disk drives. For RAID arrays it is usually the stripe width or
|
|
|
|
* the internal track size. A properly aligned multiple of
|
|
|
|
* optimal_io_size is the preferred request size for workloads where
|
|
|
|
* sustained throughput is desired.
|
2009-05-23 04:17:53 +07:00
|
|
|
*/
|
|
|
|
void blk_queue_io_opt(struct request_queue *q, unsigned int opt)
|
|
|
|
{
|
2009-09-12 02:54:52 +07:00
|
|
|
blk_limits_io_opt(&q->limits, opt);
|
2009-05-23 04:17:53 +07:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_io_opt);
|
|
|
|
|
2008-01-29 20:53:40 +07:00
|
|
|
/*
|
|
|
|
* 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(struct request_queue *t, struct request_queue *b)
|
|
|
|
{
|
2009-07-31 22:49:10 +07:00
|
|
|
blk_stack_limits(&t->limits, &b->limits, 0);
|
2009-05-23 04:17:51 +07:00
|
|
|
|
2008-05-15 06:05:54 +07:00
|
|
|
if (!t->queue_lock)
|
|
|
|
WARN_ON_ONCE(1);
|
|
|
|
else if (!test_bit(QUEUE_FLAG_CLUSTER, &b->queue_flags)) {
|
|
|
|
unsigned long flags;
|
|
|
|
spin_lock_irqsave(t->queue_lock, flags);
|
2008-04-29 19:48:33 +07:00
|
|
|
queue_flag_clear(QUEUE_FLAG_CLUSTER, t);
|
2008-05-15 06:05:54 +07:00
|
|
|
spin_unlock_irqrestore(t->queue_lock, flags);
|
|
|
|
}
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_stack_limits);
|
|
|
|
|
2009-05-23 04:17:53 +07:00
|
|
|
/**
|
|
|
|
* blk_stack_limits - adjust queue_limits for stacked devices
|
|
|
|
* @t: the stacking driver limits (top)
|
2009-06-09 11:23:22 +07:00
|
|
|
* @b: the underlying queue limits (bottom)
|
2009-05-23 04:17:53 +07:00
|
|
|
* @offset: offset to beginning of data within component device
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* Merges two queue_limit structs. Returns 0 if alignment didn't
|
|
|
|
* change. Returns -1 if adding the bottom device caused
|
|
|
|
* misalignment.
|
|
|
|
*/
|
|
|
|
int blk_stack_limits(struct queue_limits *t, struct queue_limits *b,
|
|
|
|
sector_t offset)
|
|
|
|
{
|
|
|
|
t->max_sectors = min_not_zero(t->max_sectors, b->max_sectors);
|
|
|
|
t->max_hw_sectors = min_not_zero(t->max_hw_sectors, b->max_hw_sectors);
|
2009-06-09 11:23:22 +07:00
|
|
|
t->bounce_pfn = min_not_zero(t->bounce_pfn, b->bounce_pfn);
|
2009-05-23 04:17:53 +07:00
|
|
|
|
|
|
|
t->seg_boundary_mask = min_not_zero(t->seg_boundary_mask,
|
|
|
|
b->seg_boundary_mask);
|
|
|
|
|
|
|
|
t->max_phys_segments = min_not_zero(t->max_phys_segments,
|
|
|
|
b->max_phys_segments);
|
|
|
|
|
|
|
|
t->max_hw_segments = min_not_zero(t->max_hw_segments,
|
|
|
|
b->max_hw_segments);
|
|
|
|
|
|
|
|
t->max_segment_size = min_not_zero(t->max_segment_size,
|
|
|
|
b->max_segment_size);
|
|
|
|
|
|
|
|
t->logical_block_size = max(t->logical_block_size,
|
|
|
|
b->logical_block_size);
|
|
|
|
|
|
|
|
t->physical_block_size = max(t->physical_block_size,
|
|
|
|
b->physical_block_size);
|
|
|
|
|
|
|
|
t->io_min = max(t->io_min, b->io_min);
|
|
|
|
t->no_cluster |= b->no_cluster;
|
|
|
|
|
|
|
|
/* Bottom device offset aligned? */
|
|
|
|
if (offset &&
|
|
|
|
(offset & (b->physical_block_size - 1)) != b->alignment_offset) {
|
|
|
|
t->misaligned = 1;
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* If top has no alignment offset, inherit from bottom */
|
|
|
|
if (!t->alignment_offset)
|
|
|
|
t->alignment_offset =
|
|
|
|
b->alignment_offset & (b->physical_block_size - 1);
|
|
|
|
|
|
|
|
/* Top device aligned on logical block boundary? */
|
|
|
|
if (t->alignment_offset & (t->logical_block_size - 1)) {
|
|
|
|
t->misaligned = 1;
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2009-07-31 22:49:12 +07:00
|
|
|
/* Find lcm() of optimal I/O size */
|
|
|
|
if (t->io_opt && b->io_opt)
|
|
|
|
t->io_opt = (t->io_opt * b->io_opt) / gcd(t->io_opt, b->io_opt);
|
|
|
|
else if (b->io_opt)
|
|
|
|
t->io_opt = b->io_opt;
|
|
|
|
|
|
|
|
/* Verify that optimal I/O size is a multiple of io_min */
|
|
|
|
if (t->io_min && t->io_opt % t->io_min)
|
|
|
|
return -1;
|
|
|
|
|
2009-05-23 04:17:53 +07:00
|
|
|
return 0;
|
|
|
|
}
|
2009-05-28 16:04:53 +07:00
|
|
|
EXPORT_SYMBOL(blk_stack_limits);
|
2009-05-23 04:17:53 +07:00
|
|
|
|
|
|
|
/**
|
|
|
|
* disk_stack_limits - adjust queue limits for stacked drivers
|
2009-06-09 11:23:22 +07:00
|
|
|
* @disk: MD/DM gendisk (top)
|
2009-05-23 04:17:53 +07:00
|
|
|
* @bdev: the underlying block device (bottom)
|
|
|
|
* @offset: offset to beginning of data within component device
|
|
|
|
*
|
|
|
|
* Description:
|
|
|
|
* Merges the limits for two queues. Returns 0 if alignment
|
|
|
|
* didn't change. Returns -1 if adding the bottom device caused
|
|
|
|
* misalignment.
|
|
|
|
*/
|
|
|
|
void disk_stack_limits(struct gendisk *disk, struct block_device *bdev,
|
|
|
|
sector_t offset)
|
|
|
|
{
|
|
|
|
struct request_queue *t = disk->queue;
|
|
|
|
struct request_queue *b = bdev_get_queue(bdev);
|
|
|
|
|
|
|
|
offset += get_start_sect(bdev) << 9;
|
|
|
|
|
|
|
|
if (blk_stack_limits(&t->limits, &b->limits, offset) < 0) {
|
|
|
|
char top[BDEVNAME_SIZE], bottom[BDEVNAME_SIZE];
|
|
|
|
|
|
|
|
disk_name(disk, 0, top);
|
|
|
|
bdevname(bdev, bottom);
|
|
|
|
|
|
|
|
printk(KERN_NOTICE "%s: Warning: Device %s is misaligned\n",
|
|
|
|
top, bottom);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (!t->queue_lock)
|
|
|
|
WARN_ON_ONCE(1);
|
|
|
|
else if (!test_bit(QUEUE_FLAG_CLUSTER, &b->queue_flags)) {
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
spin_lock_irqsave(t->queue_lock, flags);
|
|
|
|
if (!test_bit(QUEUE_FLAG_CLUSTER, &b->queue_flags))
|
|
|
|
queue_flag_clear(QUEUE_FLAG_CLUSTER, t);
|
|
|
|
spin_unlock_irqrestore(t->queue_lock, flags);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(disk_stack_limits);
|
|
|
|
|
2008-03-04 17:18:17 +07:00
|
|
|
/**
|
|
|
|
* blk_queue_dma_pad - set pad mask
|
|
|
|
* @q: the request queue for the device
|
|
|
|
* @mask: pad mask
|
|
|
|
*
|
2008-07-04 14:30:03 +07:00
|
|
|
* Set dma pad mask.
|
2008-03-04 17:18:17 +07:00
|
|
|
*
|
2008-07-04 14:30:03 +07:00
|
|
|
* Appending pad buffer to a request modifies the last entry of a
|
|
|
|
* scatter list such that it includes the pad buffer.
|
2008-03-04 17:18:17 +07:00
|
|
|
**/
|
|
|
|
void blk_queue_dma_pad(struct request_queue *q, unsigned int mask)
|
|
|
|
{
|
|
|
|
q->dma_pad_mask = mask;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_dma_pad);
|
|
|
|
|
2008-07-04 14:30:03 +07:00
|
|
|
/**
|
|
|
|
* blk_queue_update_dma_pad - update pad mask
|
|
|
|
* @q: the request queue for the device
|
|
|
|
* @mask: pad mask
|
|
|
|
*
|
|
|
|
* Update dma pad mask.
|
|
|
|
*
|
|
|
|
* Appending pad buffer to a request modifies the last entry of a
|
|
|
|
* scatter list such that it includes the pad buffer.
|
|
|
|
**/
|
|
|
|
void blk_queue_update_dma_pad(struct request_queue *q, unsigned int mask)
|
|
|
|
{
|
|
|
|
if (mask > q->dma_pad_mask)
|
|
|
|
q->dma_pad_mask = mask;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_update_dma_pad);
|
|
|
|
|
2008-01-29 20:53:40 +07:00
|
|
|
/**
|
|
|
|
* blk_queue_dma_drain - Set up a drain buffer for excess dma.
|
|
|
|
* @q: the request queue for the device
|
2008-02-19 17:36:53 +07:00
|
|
|
* @dma_drain_needed: fn which returns non-zero if drain is necessary
|
2008-01-29 20:53:40 +07:00
|
|
|
* @buf: physically contiguous buffer
|
|
|
|
* @size: size of the buffer in bytes
|
|
|
|
*
|
|
|
|
* Some devices have excess DMA problems and can't simply discard (or
|
|
|
|
* zero fill) the unwanted piece of the transfer. They have to have a
|
|
|
|
* real area of memory to transfer it into. The use case for this is
|
|
|
|
* ATAPI devices in DMA mode. If the packet command causes a transfer
|
|
|
|
* bigger than the transfer size some HBAs will lock up if there
|
|
|
|
* aren't DMA elements to contain the excess transfer. What this API
|
|
|
|
* does is adjust the queue so that the buf is always appended
|
|
|
|
* silently to the scatterlist.
|
|
|
|
*
|
|
|
|
* Note: This routine adjusts max_hw_segments to make room for
|
|
|
|
* appending the drain buffer. If you call
|
|
|
|
* blk_queue_max_hw_segments() or blk_queue_max_phys_segments() after
|
|
|
|
* calling this routine, you must set the limit to one fewer than your
|
|
|
|
* device can support otherwise there won't be room for the drain
|
|
|
|
* buffer.
|
|
|
|
*/
|
2008-03-04 17:30:18 +07:00
|
|
|
int blk_queue_dma_drain(struct request_queue *q,
|
2008-02-19 17:36:53 +07:00
|
|
|
dma_drain_needed_fn *dma_drain_needed,
|
|
|
|
void *buf, unsigned int size)
|
2008-01-29 20:53:40 +07:00
|
|
|
{
|
2009-05-23 04:17:50 +07:00
|
|
|
if (queue_max_hw_segments(q) < 2 || queue_max_phys_segments(q) < 2)
|
2008-01-29 20:53:40 +07:00
|
|
|
return -EINVAL;
|
|
|
|
/* make room for appending the drain */
|
2009-05-23 04:17:50 +07:00
|
|
|
blk_queue_max_hw_segments(q, queue_max_hw_segments(q) - 1);
|
|
|
|
blk_queue_max_phys_segments(q, queue_max_phys_segments(q) - 1);
|
2008-02-19 17:36:53 +07:00
|
|
|
q->dma_drain_needed = dma_drain_needed;
|
2008-01-29 20:53:40 +07:00
|
|
|
q->dma_drain_buffer = buf;
|
|
|
|
q->dma_drain_size = size;
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(blk_queue_dma_drain);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* 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(struct request_queue *q, unsigned long mask)
|
|
|
|
{
|
|
|
|
if (mask < PAGE_CACHE_SIZE - 1) {
|
|
|
|
mask = PAGE_CACHE_SIZE - 1;
|
2008-05-01 18:35:17 +07:00
|
|
|
printk(KERN_INFO "%s: set to minimum %lx\n",
|
|
|
|
__func__, mask);
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
|
2009-05-23 04:17:51 +07:00
|
|
|
q->limits.seg_boundary_mask = mask;
|
2008-01-29 20:53:40 +07:00
|
|
|
}
|
|
|
|
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:
|
2008-08-20 01:13:11 +07:00
|
|
|
* set required memory and length alignment for direct dma transactions.
|
2009-04-01 21:01:39 +07:00
|
|
|
* this is used when building direct io requests for the queue.
|
2008-01-29 20:53:40 +07:00
|
|
|
*
|
|
|
|
**/
|
|
|
|
void blk_queue_dma_alignment(struct request_queue *q, int mask)
|
|
|
|
{
|
|
|
|
q->dma_alignment = mask;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_dma_alignment);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* blk_queue_update_dma_alignment - update dma length and memory alignment
|
|
|
|
* @q: the request queue for the device
|
|
|
|
* @mask: alignment mask
|
|
|
|
*
|
|
|
|
* description:
|
2008-08-20 01:13:11 +07:00
|
|
|
* update required memory and length alignment for direct dma transactions.
|
2008-01-29 20:53:40 +07:00
|
|
|
* If the requested alignment is larger than the current alignment, then
|
|
|
|
* the current queue alignment is updated to the new value, otherwise it
|
|
|
|
* is left alone. The design of this is to allow multiple objects
|
|
|
|
* (driver, device, transport etc) to set their respective
|
|
|
|
* alignments without having them interfere.
|
|
|
|
*
|
|
|
|
**/
|
|
|
|
void blk_queue_update_dma_alignment(struct request_queue *q, int mask)
|
|
|
|
{
|
|
|
|
BUG_ON(mask > PAGE_SIZE);
|
|
|
|
|
|
|
|
if (mask > q->dma_alignment)
|
|
|
|
q->dma_alignment = mask;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(blk_queue_update_dma_alignment);
|
|
|
|
|
2008-08-28 14:27:42 +07:00
|
|
|
static int __init blk_settings_init(void)
|
2008-01-29 20:53:40 +07:00
|
|
|
{
|
|
|
|
blk_max_low_pfn = max_low_pfn - 1;
|
|
|
|
blk_max_pfn = max_pfn - 1;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
subsys_initcall(blk_settings_init);
|