unsigned int offset)
{
struct request_queue *q = bdev_get_queue(bio->bi_bdev);
- return __bio_add_page(q, bio, page, len, offset, queue_max_sectors(q));
+
+ return __bio_add_page(q, bio, page, len, offset, blk_max_size_offset(q, bio->bi_iter.bi_sector));
}
EXPORT_SYMBOL(bio_add_page);
if (unlikely(!rq))
return ERR_PTR(-ENOMEM);
+ blk_rq_set_block_pc(rq);
+
for_each_bio(bio) {
struct bio *bounce_bio = bio;
int ret;
}
EXPORT_SYMBOL(blk_make_request);
+/**
+ * blk_rq_set_block_pc - initialize a requeest to type BLOCK_PC
+ * @rq: request to be initialized
+ *
+ */
+void blk_rq_set_block_pc(struct request *rq)
+{
+ rq->cmd_type = REQ_TYPE_BLOCK_PC;
+ rq->__data_len = 0;
+ rq->__sector = (sector_t) -1;
+ rq->bio = rq->biotail = NULL;
+ memset(rq->__cmd, 0, sizeof(rq->__cmd));
+ rq->cmd = rq->__cmd;
+}
+EXPORT_SYMBOL(blk_rq_set_block_pc);
+
/**
* blk_requeue_request - put a request back on queue
* @q: request queue where request should be inserted
if (rq->errors)
err = -EIO;
+ if (rq->sense == sense) {
+ rq->sense = NULL;
+ rq->sense_len = 0;
+ }
+
return err;
}
EXPORT_SYMBOL(blk_execute_rq);
__percpu_counter_add(&q->mq_usage_counter, 1, 1000000);
smp_wmb();
- /* we have problems to freeze the queue if it's initializing */
- if (!blk_queue_bypass(q) || !blk_queue_init_done(q))
+
+ /* we have problems freezing the queue if it's initializing */
+ if (!blk_queue_dying(q) &&
+ (!blk_queue_bypass(q) || !blk_queue_init_done(q)))
return 0;
__percpu_counter_add(&q->mq_usage_counter, -1, 1000000);
RB_CLEAR_NODE(&rq->rb_node);
rq->rq_disk = NULL;
rq->part = NULL;
+ rq->start_time = jiffies;
#ifdef CONFIG_BLK_CGROUP
rq->rl = NULL;
set_start_time_ns(rq);
rq->sense = NULL;
INIT_LIST_HEAD(&rq->timeout_list);
+ rq->timeout = 0;
+
rq->end_io = NULL;
rq->end_io_data = NULL;
rq->next_rq = NULL;
if (unlikely(blk_bidi_rq(rq)))
rq->next_rq->resid_len = blk_rq_bytes(rq->next_rq);
- /*
- * Just mark start time and set the started bit. Due to memory
- * ordering, we know we'll see the correct deadline as long as
- * REQ_ATOMIC_STARTED is seen. Use the default queue timeout,
- * unless one has been set in the request.
- */
- if (!rq->timeout)
- rq->deadline = jiffies + q->rq_timeout;
- else
- rq->deadline = jiffies + rq->timeout;
+ blk_add_timer(rq);
/*
* Mark us as started and clear complete. Complete might have been
list_add_tail(&rq->queuelist, &ctx->rq_list);
blk_mq_hctx_mark_pending(hctx, ctx);
-
- /*
- * We do this early, to ensure we are on the right CPU.
- */
- blk_add_timer(rq);
}
void blk_mq_insert_request(struct request *rq, bool at_head, bool run_queue,
{
init_request_from_bio(rq, bio);
- if (blk_do_io_stat(rq)) {
- rq->start_time = jiffies;
+ if (blk_do_io_stat(rq))
blk_account_io_start(rq, 1);
- }
}
static inline bool blk_mq_merge_queue_io(struct blk_mq_hw_ctx *hctx,
blk_mq_bio_to_request(rq, bio);
blk_mq_start_request(rq, true);
- blk_add_timer(rq);
/*
* For OK queue, we are done. For error, kill it. Any other
return NOTIFY_OK;
}
+/*
+ * Alloc a tag set to be associated with one or more request queues.
+ * May fail with EINVAL for various error conditions. May adjust the
+ * requested depth down, if if it too large. In that case, the set
+ * value will be stored in set->queue_depth.
+ */
int blk_mq_alloc_tag_set(struct blk_mq_tag_set *set)
{
int i;
if (!set->nr_hw_queues)
return -EINVAL;
- if (!set->queue_depth || set->queue_depth > BLK_MQ_MAX_DEPTH)
+ if (!set->queue_depth)
return -EINVAL;
if (set->queue_depth < set->reserved_tags + BLK_MQ_TAG_MIN)
return -EINVAL;
if (!set->nr_hw_queues || !set->ops->queue_rq || !set->ops->map_queue)
return -EINVAL;
+ if (set->queue_depth > BLK_MQ_MAX_DEPTH) {
+ pr_info("blk-mq: reduced tag depth to %u\n",
+ BLK_MQ_MAX_DEPTH);
+ set->queue_depth = BLK_MQ_MAX_DEPTH;
+ }
set->tags = kmalloc_node(set->nr_hw_queues *
sizeof(struct blk_mq_tags *),
lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
lim->max_segment_size = BLK_MAX_SEGMENT_SIZE;
lim->max_sectors = lim->max_hw_sectors = BLK_SAFE_MAX_SECTORS;
+ lim->chunk_sectors = 0;
lim->max_write_same_sectors = 0;
lim->max_discard_sectors = 0;
lim->discard_granularity = 0;
}
EXPORT_SYMBOL(blk_queue_max_hw_sectors);
+/**
+ * blk_queue_chunk_sectors - set size of the chunk for this queue
+ * @q: the request queue for the device
+ * @chunk_sectors: chunk sectors in the usual 512b unit
+ *
+ * Description:
+ * If a driver doesn't want IOs to cross a given chunk size, it can set
+ * this limit and prevent merging across chunks. Note that the chunk size
+ * must currently be a power-of-2 in sectors.
+ **/
+void blk_queue_chunk_sectors(struct request_queue *q, unsigned int chunk_sectors)
+{
+ BUG_ON(!is_power_of_2(chunk_sectors));
+ q->limits.chunk_sectors = chunk_sectors;
+}
+EXPORT_SYMBOL(blk_queue_chunk_sectors);
+
/**
* blk_queue_max_discard_sectors - set max sectors for a single discard
* @q: the request queue for the device
* fill in request structure
*/
rq->cmd_len = hdr->request_len;
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
rq->timeout = msecs_to_jiffies(hdr->timeout);
if (!rq->timeout)
rq = blk_get_request(q, rw, GFP_KERNEL);
if (!rq)
return ERR_PTR(-ENOMEM);
+ blk_rq_set_block_pc(rq);
+
ret = blk_fill_sgv4_hdr_rq(q, rq, hdr, bd, has_write_perm);
if (ret)
goto out;
* fill in request structure
*/
rq->cmd_len = hdr->cmd_len;
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
rq->timeout = msecs_to_jiffies(hdr->timeout);
if (!rq->timeout)
rq = blk_get_request(q, writing ? WRITE : READ, GFP_KERNEL);
if (!rq)
return -ENOMEM;
+ blk_rq_set_block_pc(rq);
if (blk_fill_sghdr_rq(q, rq, hdr, mode)) {
blk_put_request(rq);
memset(sense, 0, sizeof(sense));
rq->sense = sense;
rq->sense_len = 0;
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
+ blk_rq_set_block_pc(rq);
blk_execute_rq(q, disk, rq, 0);
int err;
rq = blk_get_request(q, WRITE, __GFP_WAIT);
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
+ blk_rq_set_block_pc(rq);
rq->timeout = BLK_DEFAULT_SG_TIMEOUT;
rq->cmd[0] = cmd;
rq->cmd[4] = data;
#include <../drivers/ata/ahci.h>
#include <linux/export.h>
#include <linux/debugfs.h>
+#include <linux/prefetch.h>
#include "mtip32xx.h"
#define HW_CMD_SLOT_SZ (MTIP_MAX_COMMAND_SLOTS * 32)
/* Map the scatter list for DMA access */
nents = dma_map_sg(&dd->pdev->dev, command->sg, nents, dma_dir);
+ prefetch(&port->flags);
+
command->scatter_ents = nents;
/*
fis = command->command;
fis->type = 0x27;
fis->opts = 1 << 7;
- if (rq_data_dir(rq) == READ)
+ if (dma_dir == DMA_FROM_DEVICE)
fis->command = ATA_CMD_FPDMA_READ;
else
fis->command = ATA_CMD_FPDMA_WRITE;
fis->res3 = 0;
fill_command_sg(dd, command, nents);
- if (command->unaligned)
+ if (unlikely(command->unaligned))
fis->device |= 1 << 7;
/* Populate the command header */
* To prevent this command from being issued
* if an internal command is in progress or error handling is active.
*/
- if (port->flags & MTIP_PF_PAUSE_IO) {
+ if (unlikely(port->flags & MTIP_PF_PAUSE_IO)) {
set_bit(rq->tag, port->cmds_to_issue);
set_bit(MTIP_PF_ISSUE_CMDS_BIT, &port->flags);
return;
struct driver_data *dd = hctx->queue->queuedata;
struct mtip_cmd *cmd = blk_mq_rq_to_pdu(rq);
- if (!dd->unal_qdepth || rq_data_dir(rq) == READ)
+ if (rq_data_dir(rq) == READ || !dd->unal_qdepth)
return false;
/*
{
int ret;
- if (mtip_check_unal_depth(hctx, rq))
+ if (unlikely(mtip_check_unal_depth(hctx, rq)))
return BLK_MQ_RQ_QUEUE_BUSY;
ret = mtip_submit_request(hctx, rq);
- if (!ret)
+ if (likely(!ret))
return BLK_MQ_RQ_QUEUE_OK;
rq->errors = ret;
struct workqueue_struct *isr_workq;
- struct mtip_work work[MTIP_MAX_SLOT_GROUPS];
-
atomic_t irq_workers_active;
+ struct mtip_work work[MTIP_MAX_SLOT_GROUPS];
+
int isr_binding;
struct block_device *bdev;
- int unal_qdepth; /* qdepth of unaligned IO queue */
-
struct list_head online_list; /* linkage for online list */
struct list_head remove_list; /* linkage for removing list */
+
+ int unal_qdepth; /* qdepth of unaligned IO queue */
};
#endif
rq = blk_get_request(q, (cgc->data_direction == CGC_DATA_WRITE) ?
WRITE : READ, __GFP_WAIT);
+ blk_rq_set_block_pc(rq);
if (cgc->buflen) {
ret = blk_rq_map_kern(q, rq, cgc->buffer, cgc->buflen,
memcpy(rq->cmd, cgc->cmd, CDROM_PACKET_SIZE);
rq->timeout = 60*HZ;
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
if (cgc->quiet)
rq->cmd_flags |= REQ_QUIET;
ret = -ENOMEM;
break;
}
+ blk_rq_set_block_pc(rq);
ret = blk_rq_map_user(q, rq, NULL, ubuf, len, GFP_KERNEL);
if (ret) {
rq->cmd[9] = 0xf8;
rq->cmd_len = 12;
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
rq->timeout = 60 * HZ;
bio = rq->bio;
"%s: blk_get_request failed\n", __func__);
return NULL;
}
+ blk_rq_set_block_pc(rq);
if (buflen && blk_rq_map_kern(q, rq, buffer, buflen, GFP_NOIO)) {
blk_put_request(rq);
return NULL;
}
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
REQ_FAILFAST_DRIVER;
rq->retries = ALUA_FAILOVER_RETRIES;
return NULL;
}
+ blk_rq_set_block_pc(rq);
rq->cmd_len = COMMAND_SIZE(cmd);
rq->cmd[0] = cmd;
break;
}
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
REQ_FAILFAST_DRIVER;
rq->timeout = CLARIION_TIMEOUT;
if (!req)
return SCSI_DH_RES_TEMP_UNAVAIL;
- req->cmd_type = REQ_TYPE_BLOCK_PC;
+ blk_rq_set_block_pc(req);
req->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
REQ_FAILFAST_DRIVER;
req->cmd_len = COMMAND_SIZE(TEST_UNIT_READY);
if (!req)
return SCSI_DH_RES_TEMP_UNAVAIL;
- req->cmd_type = REQ_TYPE_BLOCK_PC;
+ blk_rq_set_block_pc(req);
req->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
REQ_FAILFAST_DRIVER;
req->cmd_len = COMMAND_SIZE(START_STOP);
"get_rdac_req: blk_get_request failed.\n");
return NULL;
}
+ blk_rq_set_block_pc(rq);
if (buflen && blk_rq_map_kern(q, rq, buffer, buflen, GFP_NOIO)) {
blk_put_request(rq);
return NULL;
}
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT |
REQ_FAILFAST_DRIVER;
rq->retries = RDAC_RETRIES;
if (unlikely(!req))
return ERR_PTR(-ENOMEM);
+ blk_rq_set_block_pc(req);
return req;
}
}
}
or->request = req;
- req->cmd_type = REQ_TYPE_BLOCK_PC;
req->cmd_flags |= REQ_QUIET;
req->timeout = or->timeout;
ret = PTR_ERR(req);
goto out;
}
- req->cmd_type = REQ_TYPE_BLOCK_PC;
+ blk_rq_set_block_pc(req);
or->in.req = or->request->next_rq = req;
}
} else if (has_in)
if (!req)
return DRIVER_ERROR << 24;
- req->cmd_type = REQ_TYPE_BLOCK_PC;
+ blk_rq_set_block_pc(req);
req->cmd_flags |= REQ_QUIET;
SRpnt->bio = NULL;
*/
req = blk_get_request(sdev->request_queue, READ, GFP_KERNEL);
+ blk_rq_set_block_pc(req);
+
req->cmd[0] = ALLOW_MEDIUM_REMOVAL;
req->cmd[1] = 0;
req->cmd[2] = 0;
req->cmd_len = COMMAND_SIZE(req->cmd[0]);
- req->cmd_type = REQ_TYPE_BLOCK_PC;
req->cmd_flags |= REQ_QUIET;
req->timeout = 10 * HZ;
req->retries = 5;
req = blk_get_request(sdev->request_queue, write, __GFP_WAIT);
if (!req)
return ret;
+ blk_rq_set_block_pc(req);
if (bufflen && blk_rq_map_kern(sdev->request_queue, req,
buffer, bufflen, __GFP_WAIT))
req->sense_len = 0;
req->retries = retries;
req->timeout = timeout;
- req->cmd_type = REQ_TYPE_BLOCK_PC;
req->cmd_flags |= flags | REQ_QUIET | REQ_PREEMPT;
/*
if (!rq)
return -ENOMEM;
+ blk_rq_set_block_pc(rq);
memcpy(rq->cmd, cmd, hp->cmd_len);
-
rq->cmd_len = hp->cmd_len;
- rq->cmd_type = REQ_TYPE_BLOCK_PC;
srp->rq = rq;
rq->end_io_data = srp;
if (!req)
return DRIVER_ERROR << 24;
- req->cmd_type = REQ_TYPE_BLOCK_PC;
+ blk_rq_set_block_pc(req);
req->cmd_flags |= REQ_QUIET;
mdata->null_mapped = 1;
ret = TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
goto fail;
}
+
+ blk_rq_set_block_pc(req);
} else {
BUG_ON(!cmd->data_length);
}
}
- req->cmd_type = REQ_TYPE_BLOCK_PC;
req->end_io = pscsi_req_done;
req->end_io_data = cmd;
req->cmd_len = scsi_command_size(pt->pscsi_cdb);
BLK_MQ_S_STOPPED = 0,
BLK_MQ_S_TAG_ACTIVE = 1,
- BLK_MQ_MAX_DEPTH = 2048,
+ BLK_MQ_MAX_DEPTH = 10240,
BLK_MQ_CPU_WORK_BATCH = 8,
};
unsigned long seg_boundary_mask;
unsigned int max_hw_sectors;
+ unsigned int chunk_sectors;
unsigned int max_sectors;
unsigned int max_segment_size;
unsigned int physical_block_size;
extern struct request *blk_get_request(struct request_queue *, int, gfp_t);
extern struct request *blk_make_request(struct request_queue *, struct bio *,
gfp_t);
+extern void blk_rq_set_block_pc(struct request *);
extern void blk_requeue_request(struct request_queue *, struct request *);
extern void blk_add_request_payload(struct request *rq, struct page *page,
unsigned int len);
return q->limits.max_sectors;
}
+/*
+ * Return maximum size of a request at given offset. Only valid for
+ * file system requests.
+ */
+static inline unsigned int blk_max_size_offset(struct request_queue *q,
+ sector_t offset)
+{
+ if (!q->limits.chunk_sectors)
+ return q->limits.max_hw_sectors;
+
+ return q->limits.chunk_sectors -
+ (offset & (q->limits.chunk_sectors - 1));
+}
+
static inline unsigned int blk_rq_get_max_sectors(struct request *rq)
{
struct request_queue *q = rq->q;
if (unlikely(rq->cmd_type == REQ_TYPE_BLOCK_PC))
return q->limits.max_hw_sectors;
- return blk_queue_get_max_sectors(q, rq->cmd_flags);
+ if (!q->limits.chunk_sectors)
+ return blk_queue_get_max_sectors(q, rq->cmd_flags);
+
+ return min(blk_max_size_offset(q, blk_rq_pos(rq)),
+ blk_queue_get_max_sectors(q, rq->cmd_flags));
}
static inline unsigned int blk_rq_count_bios(struct request *rq)
extern void blk_queue_bounce_limit(struct request_queue *, u64);
extern void blk_limits_max_hw_sectors(struct queue_limits *, unsigned int);
extern void blk_queue_max_hw_sectors(struct request_queue *, unsigned int);
+extern void blk_queue_chunk_sectors(struct request_queue *, unsigned int);
extern void blk_queue_max_segments(struct request_queue *, unsigned short);
extern void blk_queue_max_segment_size(struct request_queue *, unsigned int);
extern void blk_queue_max_discard_sectors(struct request_queue *q,