Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull s390 updates from Martin Schwidefsky:
 "The most notable change for this pull request is the ftrace rework
  from Heiko.  It brings a small performance improvement and the ground
  work to support a new gcc option to replace the mcount blocks with a
  single nop.

  Two new s390 specific system calls are added to emulate user space
  mmio for PCI, an artifact of the how PCI memory is accessed.

  Two patches for the memory management with changes to common code.
  For KVM mm_forbids_zeropage is added which disables the empty zero
  page for an mm that is used by a KVM process.  And an optimization,
  pmdp_get_and_clear_full is added analog to ptep_get_and_clear_full.

  Some micro optimization for the cmpxchg and the spinlock code.

  And as usual bug fixes and cleanups"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (46 commits)
  s390/cputime: fix 31-bit compile
  s390/scm_block: make the number of reqs per HW req configurable
  s390/scm_block: handle multiple requests in one HW request
  s390/scm_block: allocate aidaw pages only when necessary
  s390/scm_block: use mempool to manage aidaw requests
  s390/eadm: change timeout value
  s390/mm: fix memory leak of ptlock in pmd_free_tlb
  s390: use local symbol names in entry[64].S
  s390/ptrace: always include vector registers in core files
  s390/simd: clear vector register pointer on fork/clone
  s390: translate cputime magic constants to macros
  s390/idle: convert open coded idle time seqcount
  s390/idle: add missing irq off lockdep annotation
  s390/debug: avoid function call for debug_sprintf_*
  s390/kprobes: fix instruction copy for out of line execution
  s390: remove diag 44 calls from cpu_relax()
  s390/dasd: retry partition detection
  s390/dasd: fix list corruption for sleep_on requests
  s390/dasd: fix infinite term I/O loop
  s390/dasd: remove unused code
  ...

1 files changed, 47 insertions, 22 deletions

diff --git a/drivers/s390/block/scm_blk_cluster.c b/drivers/s390/block/scm_blk_cluster.c
index 9aae909d47a5..09db45296eed 100644
--- a/drivers/s390/block/scm_blk_cluster.c
+++ b/drivers/s390/block/scm_blk_cluster.c
@@ -57,39 +57,52 @@ void scm_request_cluster_init(struct scm_request *scmrq)
 	scmrq->cluster.state = CLUSTER_NONE;
 }
 
-static bool clusters_intersect(struct scm_request *A, struct scm_request *B)
+static bool clusters_intersect(struct request *A, struct request *B)
 {
 	unsigned long firstA, lastA, firstB, lastB;
 
-	firstA = ((u64) blk_rq_pos(A->request) << 9) / CLUSTER_SIZE;
-	lastA = (((u64) blk_rq_pos(A->request) << 9) +
-		    blk_rq_bytes(A->request) - 1) / CLUSTER_SIZE;
+	firstA = ((u64) blk_rq_pos(A) << 9) / CLUSTER_SIZE;
+	lastA = (((u64) blk_rq_pos(A) << 9) +
+		    blk_rq_bytes(A) - 1) / CLUSTER_SIZE;
 
-	firstB = ((u64) blk_rq_pos(B->request) << 9) / CLUSTER_SIZE;
-	lastB = (((u64) blk_rq_pos(B->request) << 9) +
-		    blk_rq_bytes(B->request) - 1) / CLUSTER_SIZE;
+	firstB = ((u64) blk_rq_pos(B) << 9) / CLUSTER_SIZE;
+	lastB = (((u64) blk_rq_pos(B) << 9) +
+		    blk_rq_bytes(B) - 1) / CLUSTER_SIZE;
 
 	return (firstB <= lastA && firstA <= lastB);
 }
 
 bool scm_reserve_cluster(struct scm_request *scmrq)
 {
+	struct request *req = scmrq->request[scmrq->aob->request.msb_count];
 	struct scm_blk_dev *bdev = scmrq->bdev;
 	struct scm_request *iter;
+	int pos, add = 1;
 
 	if (write_cluster_size == 0)
 		return true;
 
 	spin_lock(&bdev->lock);
 	list_for_each_entry(iter, &bdev->cluster_list, cluster.list) {
-		if (clusters_intersect(scmrq, iter) &&
-		    (rq_data_dir(scmrq->request) == WRITE ||
-		     rq_data_dir(iter->request) == WRITE)) {
-			spin_unlock(&bdev->lock);
-			return false;
+		if (iter == scmrq) {
+			/*
+			 * We don't have to use clusters_intersect here, since
+			 * cluster requests are always started separately.
+			 */
+			add = 0;
+			continue;
+		}
+		for (pos = 0; pos <= iter->aob->request.msb_count; pos++) {
+			if (clusters_intersect(req, iter->request[pos]) &&
+			    (rq_data_dir(req) == WRITE ||
+			     rq_data_dir(iter->request[pos]) == WRITE)) {
+				spin_unlock(&bdev->lock);
+				return false;
+			}
 		}
 	}
-	list_add(&scmrq->cluster.list, &bdev->cluster_list);
+	if (add)
+		list_add(&scmrq->cluster.list, &bdev->cluster_list);
 	spin_unlock(&bdev->lock);
 
 	return true;
@@ -114,14 +127,14 @@ void scm_blk_dev_cluster_setup(struct scm_blk_dev *bdev)
 	blk_queue_io_opt(bdev->rq, CLUSTER_SIZE);
 }
 
-static void scm_prepare_cluster_request(struct scm_request *scmrq)
+static int scm_prepare_cluster_request(struct scm_request *scmrq)
 {
 	struct scm_blk_dev *bdev = scmrq->bdev;
 	struct scm_device *scmdev = bdev->gendisk->private_data;
-	struct request *req = scmrq->request;
-	struct aidaw *aidaw = scmrq->aidaw;
+	struct request *req = scmrq->request[0];
 	struct msb *msb = &scmrq->aob->msb[0];
 	struct req_iterator iter;
+	struct aidaw *aidaw;
 	struct bio_vec bv;
 	int i = 0;
 	u64 addr;
@@ -131,11 +144,9 @@ static void scm_prepare_cluster_request(struct scm_request *scmrq)
 		scmrq->cluster.state = CLUSTER_READ;
 		/* fall through */
 	case CLUSTER_READ:
-		scmrq->aob->request.msb_count = 1;
 		msb->bs = MSB_BS_4K;
 		msb->oc = MSB_OC_READ;
 		msb->flags = MSB_FLAG_IDA;
-		msb->data_addr = (u64) aidaw;
 		msb->blk_count = write_cluster_size;
 
 		addr = scmdev->address + ((u64) blk_rq_pos(req) << 9);
@@ -146,6 +157,12 @@ static void scm_prepare_cluster_request(struct scm_request *scmrq)
 			       CLUSTER_SIZE))
 			msb->blk_count = 2 * write_cluster_size;
 
+		aidaw = scm_aidaw_fetch(scmrq, msb->blk_count * PAGE_SIZE);
+		if (!aidaw)
+			return -ENOMEM;
+
+		scmrq->aob->request.msb_count = 1;
+		msb->data_addr = (u64) aidaw;
 		for (i = 0; i < msb->blk_count; i++) {
 			aidaw->data_addr = (u64) scmrq->cluster.buf[i];
 			aidaw++;
@@ -153,6 +170,7 @@ static void scm_prepare_cluster_request(struct scm_request *scmrq)
 
 		break;
 	case CLUSTER_WRITE:
+		aidaw = (void *) msb->data_addr;
 		msb->oc = MSB_OC_WRITE;
 
 		for (addr = msb->scm_addr;
@@ -173,22 +191,29 @@ static void scm_prepare_cluster_request(struct scm_request *scmrq)
 		}
 		break;
 	}
+	return 0;
 }
 
 bool scm_need_cluster_request(struct scm_request *scmrq)
 {
-	if (rq_data_dir(scmrq->request) == READ)
+	int pos = scmrq->aob->request.msb_count;
+
+	if (rq_data_dir(scmrq->request[pos]) == READ)
 		return false;
 
-	return blk_rq_bytes(scmrq->request) < CLUSTER_SIZE;
+	return blk_rq_bytes(scmrq->request[pos]) < CLUSTER_SIZE;
 }
 
 /* Called with queue lock held. */
 void scm_initiate_cluster_request(struct scm_request *scmrq)
 {
-	scm_prepare_cluster_request(scmrq);
+	if (scm_prepare_cluster_request(scmrq))
+		goto requeue;
 	if (eadm_start_aob(scmrq->aob))
-		scm_request_requeue(scmrq);
+		goto requeue;
+	return;
+requeue:
+	scm_request_requeue(scmrq);
 }
 
 bool scm_test_cluster_request(struct scm_request *scmrq)