#include <linux/slab.h>
#include <linux/swap.h>
#include <linux/writeback.h>
-#include <linux/blkdev.h>
+#include <linux/interrupt.h>
+#include <linux/cpu.h>
/*
* for max sense size
/*
* Controlling structure to kblockd
*/
-static struct workqueue_struct *kblockd_workqueue;
+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);
+static DEFINE_PER_CPU(struct list_head, blk_cpu_done);
+
/* Amount of time in which a process may batch requests */
#define BLK_BATCH_TIME (HZ/50UL)
EXPORT_SYMBOL(blk_queue_merge_bvec);
+void blk_queue_softirq_done(request_queue_t *q, softirq_done_fn *fn)
+{
+ q->softirq_done_fn = fn;
+}
+
+EXPORT_SYMBOL(blk_queue_softirq_done);
+
/**
* blk_queue_make_request - define an alternate make_request function for a device
* @q: the request queue for the device to be affected
static inline void rq_init(request_queue_t *q, struct request *rq)
{
INIT_LIST_HEAD(&rq->queuelist);
+ INIT_LIST_HEAD(&rq->donelist);
rq->errors = 0;
rq->rq_status = RQ_ACTIVE;
rq->sense = NULL;
rq->end_io = NULL;
rq->end_io_data = NULL;
+ rq->completion_data = NULL;
}
/**
* blk_queue_ordered - does this queue support ordered writes
* @q: the request queue
* @ordered: one of QUEUE_ORDERED_*
+ * @prepare_flush_fn: rq setup helper for cache flush ordered writes
*
* Description:
* For journalled file systems, doing ordered writes on a commit
return -EINVAL;
}
+ q->ordered = ordered;
q->next_ordered = ordered;
q->prepare_flush_fn = prepare_flush_fn;
rq->end_io = end_io;
q->prepare_flush_fn(q, rq);
- __elv_add_request(q, rq, ELEVATOR_INSERT_FRONT, 0);
+ elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
}
static inline struct request *start_ordered(request_queue_t *q,
else
q->ordseq |= QUEUE_ORDSEQ_POSTFLUSH;
- __elv_add_request(q, rq, ELEVATOR_INSERT_FRONT, 0);
+ elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
if (q->ordered & QUEUE_ORDERED_PREFLUSH) {
queue_flush(q, QUEUE_ORDERED_PREFLUSH);
int blk_do_ordered(request_queue_t *q, struct request **rqp)
{
- struct request *rq = *rqp, *allowed_rq;
+ struct request *rq = *rqp;
int is_barrier = blk_fs_request(rq) && blk_barrier_rq(rq);
if (!q->ordseq) {
}
}
+ /*
+ * Ordered sequence in progress
+ */
+
+ /* Special requests are not subject to ordering rules. */
+ if (!blk_fs_request(rq) &&
+ rq != &q->pre_flush_rq && rq != &q->post_flush_rq)
+ return 1;
+
if (q->ordered & QUEUE_ORDERED_TAG) {
+ /* Ordered by tag. Blocking the next barrier is enough. */
if (is_barrier && rq != &q->bar_rq)
*rqp = NULL;
- return 1;
- }
-
- switch (blk_ordered_cur_seq(q)) {
- case QUEUE_ORDSEQ_PREFLUSH:
- allowed_rq = &q->pre_flush_rq;
- break;
- case QUEUE_ORDSEQ_BAR:
- allowed_rq = &q->bar_rq;
- break;
- case QUEUE_ORDSEQ_POSTFLUSH:
- allowed_rq = &q->post_flush_rq;
- break;
- default:
- allowed_rq = NULL;
- break;
+ } else {
+ /* Ordered by draining. Wait for turn. */
+ WARN_ON(blk_ordered_req_seq(rq) < blk_ordered_cur_seq(q));
+ if (blk_ordered_req_seq(rq) > blk_ordered_cur_seq(q))
+ *rqp = NULL;
}
- if (rq != allowed_rq &&
- (blk_fs_request(rq) || rq == &q->pre_flush_rq ||
- rq == &q->post_flush_rq))
- *rqp = NULL;
-
return 1;
}
* 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.
+ * buffers for doing I/O to pages residing above @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);
+ int dma = 0;
+
+ q->bounce_gfp = GFP_NOIO;
+#if BITS_PER_LONG == 64
+ /* 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 (bounce_pfn < (0xffffffff>>PAGE_SHIFT))
+ dma = 1;
+ q->bounce_pfn = max_low_pfn;
+#else
+ if (bounce_pfn < blk_max_low_pfn)
+ dma = 1;
+ q->bounce_pfn = bounce_pfn;
+#endif
+ if (dma) {
init_emergency_isa_pool();
q->bounce_gfp = GFP_NOIO | GFP_DMA;
- } else
- q->bounce_gfp = GFP_NOIO;
-
- q->bounce_pfn = bounce_pfn;
+ q->bounce_pfn = bounce_pfn;
+ }
}
EXPORT_SYMBOL(blk_queue_bounce_limit);
* 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)
+void blk_queue_max_sectors(request_queue_t *q, unsigned int max_sectors)
{
if ((max_sectors << 9) < PAGE_CACHE_SIZE) {
max_sectors = 1 << (PAGE_CACHE_SHIFT - 9);
disk->stamp = now;
}
+EXPORT_SYMBOL_GPL(disk_round_stats);
+
/*
* queue lock must be held
*/
/**
* blk_end_sync_rq - executes a completion event on a request
* @rq: request to complete
+ * @error: end io status of the request
*/
void blk_end_sync_rq(struct request *rq, int error)
{
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 void init_request_from_bio(struct request *req, struct bio *bio)
{
req->flags |= REQ_CMD;
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);
+ disk_stat_add(req->rq_disk, sectors[rw], nr_bytes >> 9);
}
total_bytes = bio_nbytes = 0;
EXPORT_SYMBOL(end_that_request_chunk);
+/*
+ * splice the completion data to a local structure and hand off to
+ * process_completion_queue() to complete the requests
+ */
+static void blk_done_softirq(struct softirq_action *h)
+{
+ struct list_head *cpu_list;
+ LIST_HEAD(local_list);
+
+ local_irq_disable();
+ cpu_list = &__get_cpu_var(blk_cpu_done);
+ list_splice_init(cpu_list, &local_list);
+ local_irq_enable();
+
+ while (!list_empty(&local_list)) {
+ struct request *rq = list_entry(local_list.next, struct request, donelist);
+
+ list_del_init(&rq->donelist);
+ rq->q->softirq_done_fn(rq);
+ }
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+static int blk_cpu_notify(struct notifier_block *self, unsigned long action,
+ void *hcpu)
+{
+ /*
+ * If a CPU goes away, splice its entries to the current CPU
+ * and trigger a run of the softirq
+ */
+ if (action == CPU_DEAD) {
+ int cpu = (unsigned long) hcpu;
+
+ local_irq_disable();
+ list_splice_init(&per_cpu(blk_cpu_done, cpu),
+ &__get_cpu_var(blk_cpu_done));
+ raise_softirq_irqoff(BLOCK_SOFTIRQ);
+ local_irq_enable();
+ }
+
+ return NOTIFY_OK;
+}
+
+
+static struct notifier_block __devinitdata blk_cpu_notifier = {
+ .notifier_call = blk_cpu_notify,
+};
+
+#endif /* CONFIG_HOTPLUG_CPU */
+
+/**
+ * blk_complete_request - end I/O on a request
+ * @req: the request being processed
+ *
+ * Description:
+ * Ends all I/O on a request. It does not handle partial completions,
+ * unless the driver actually implements this in its completionc callback
+ * through requeueing. Theh actual completion happens out-of-order,
+ * through a softirq handler. The user must have registered a completion
+ * callback through blk_queue_softirq_done().
+ **/
+
+void blk_complete_request(struct request *req)
+{
+ struct list_head *cpu_list;
+ unsigned long flags;
+
+ BUG_ON(!req->q->softirq_done_fn);
+
+ local_irq_save(flags);
+
+ cpu_list = &__get_cpu_var(blk_cpu_done);
+ list_add_tail(&req->donelist, cpu_list);
+ raise_softirq_irqoff(BLOCK_SOFTIRQ);
+
+ local_irq_restore(flags);
+}
+
+EXPORT_SYMBOL(blk_complete_request);
+
/*
* queue lock must be held
*/
int __init blk_dev_init(void)
{
+ int i;
+
kblockd_workqueue = create_workqueue("kblockd");
if (!kblockd_workqueue)
panic("Failed to create kblockd\n");
iocontext_cachep = kmem_cache_create("blkdev_ioc",
sizeof(struct io_context), 0, SLAB_PANIC, NULL, NULL);
+ for_each_cpu(i)
+ INIT_LIST_HEAD(&per_cpu(blk_cpu_done, i));
+
+ open_softirq(BLOCK_SOFTIRQ, blk_done_softirq, NULL);
+#ifdef CONFIG_HOTPLUG_CPU
+ register_cpu_notifier(&blk_cpu_notifier);
+#endif
+
blk_max_low_pfn = max_low_pfn;
blk_max_pfn = max_pfn;
projects / linux-2.6-omap-h63xx.git / blobdiff