[S390] dasd: add hyper PAV support to DASD device driver, part 1

Parallel access volumes (PAV) is a storage server feature, that allows
to start multiple channel programs on the same DASD in parallel. It
defines alias devices which can be used as alternative paths to the
same disk. With the old base PAV support we only needed rudimentary
functionality in the DASD device driver. As the mapping between base
and alias devices was static, we just had to export an identifier
(uid) and could leave the combining of devices to external layers
like a device mapper multipath.
Now hyper PAV removes the requirement to dedicate alias devices to
specific base devices. Instead each alias devices can be combined with
multiple base device on a per request basis. This requires full
support by the DASD device driver as now each channel program itself
has to identify the target base device.
The changes to the dasd device driver and the ECKD discipline are:
- Separate subchannel device representation (dasd_device) from block
  device representation (dasd_block). Only base devices are block
  devices.
- Gather information about base and alias devices and possible
  combinations.
- For each request decide which dasd_device should be used (base or
  alias) and build specific channel program.
- Support summary unit checks, which allow the storage server to
  upgrade / downgrade between base and hyper PAV at runtime (support
  is mandatory).

Signed-off-by: Stefan Weinhuber <wein@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

1 files changed, 60 insertions, 47 deletions

diff --git a/drivers/s390/block/dasd_diag.c b/drivers/s390/block/dasd_diag.c
index 571320ab..d91df38 100644
--- a/drivers/s390/block/dasd_diag.c
+++ b/drivers/s390/block/dasd_diag.c
@@ -142,7 +142,7 @@ dasd_diag_erp(struct dasd_device *device)
 	int rc;
 
 	mdsk_term_io(device);
-	rc = mdsk_init_io(device, device->bp_block, 0, NULL);
+	rc = mdsk_init_io(device, device->block->bp_block, 0, NULL);
 	if (rc)
 		DEV_MESSAGE(KERN_WARNING, device, "DIAG ERP unsuccessful, "
 			    "rc=%d", rc);
@@ -158,11 +158,11 @@ dasd_start_diag(struct dasd_ccw_req * cqr)
 	struct dasd_diag_req *dreq;
 	int rc;
 
-	device = cqr->device;
+	device = cqr->startdev;
 	if (cqr->retries < 0) {
 		DEV_MESSAGE(KERN_WARNING, device, "DIAG start_IO: request %p "
 			    "- no retry left)", cqr);
-		cqr->status = DASD_CQR_FAILED;
+		cqr->status = DASD_CQR_ERROR;
 		return -EIO;
 	}
 	private = (struct dasd_diag_private *) device->private;
@@ -184,7 +184,7 @@ dasd_start_diag(struct dasd_ccw_req * cqr)
 	switch (rc) {
 	case 0: /* Synchronous I/O finished successfully */
 		cqr->stopclk = get_clock();
-		cqr->status = DASD_CQR_DONE;
+		cqr->status = DASD_CQR_SUCCESS;
 		/* Indicate to calling function that only a dasd_schedule_bh()
 		   and no timer is needed */
                 rc = -EACCES;
@@ -209,12 +209,12 @@ dasd_diag_term_IO(struct dasd_ccw_req * cqr)
 {
 	struct dasd_device *device;
 
-	device = cqr->device;
+	device = cqr->startdev;
 	mdsk_term_io(device);
-	mdsk_init_io(device, device->bp_block, 0, NULL);
-	cqr->status = DASD_CQR_CLEAR;
+	mdsk_init_io(device, device->block->bp_block, 0, NULL);
+	cqr->status = DASD_CQR_CLEAR_PENDING;
 	cqr->stopclk = get_clock();
-	dasd_schedule_bh(device);
+	dasd_schedule_device_bh(device);
 	return 0;
 }
 
@@ -247,7 +247,7 @@ dasd_ext_handler(__u16 code)
 		return;
 	}
 	cqr = (struct dasd_ccw_req *) ip;
-	device = (struct dasd_device *) cqr->device;
+	device = (struct dasd_device *) cqr->startdev;
 	if (strncmp(device->discipline->ebcname, (char *) &cqr->magic, 4)) {
 		DEV_MESSAGE(KERN_WARNING, device,
 			    " magic number of dasd_ccw_req 0x%08X doesn't"
@@ -260,10 +260,10 @@ dasd_ext_handler(__u16 code)
 	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
 
 	/* Check for a pending clear operation */
-	if (cqr->status == DASD_CQR_CLEAR) {
-		cqr->status = DASD_CQR_QUEUED;
-		dasd_clear_timer(device);
-		dasd_schedule_bh(device);
+	if (cqr->status == DASD_CQR_CLEAR_PENDING) {
+		cqr->status = DASD_CQR_CLEARED;
+		dasd_device_clear_timer(device);
+		dasd_schedule_device_bh(device);
 		spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
 		return;
 	}
@@ -272,11 +272,11 @@ dasd_ext_handler(__u16 code)
 
 	expires = 0;
 	if (status == 0) {
-		cqr->status = DASD_CQR_DONE;
+		cqr->status = DASD_CQR_SUCCESS;
 		/* Start first request on queue if possible -> fast_io. */
 		if (!list_empty(&device->ccw_queue)) {
 			next = list_entry(device->ccw_queue.next,
-					  struct dasd_ccw_req, list);
+					  struct dasd_ccw_req, devlist);
 			if (next->status == DASD_CQR_QUEUED) {
 				rc = dasd_start_diag(next);
 				if (rc == 0)
@@ -296,10 +296,10 @@ dasd_ext_handler(__u16 code)
 	}
 
 	if (expires != 0)
-		dasd_set_timer(device, expires);
+		dasd_device_set_timer(device, expires);
 	else
-		dasd_clear_timer(device);
-	dasd_schedule_bh(device);
+		dasd_device_clear_timer(device);
+	dasd_schedule_device_bh(device);
 
 	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
 }
@@ -309,6 +309,7 @@ dasd_ext_handler(__u16 code)
 static int
 dasd_diag_check_device(struct dasd_device *device)
 {
+	struct dasd_block *block;
 	struct dasd_diag_private *private;
 	struct dasd_diag_characteristics *rdc_data;
 	struct dasd_diag_bio bio;
@@ -328,6 +329,16 @@ dasd_diag_check_device(struct dasd_device *device)
 		ccw_device_get_id(device->cdev, &private->dev_id);
 		device->private = (void *) private;
 	}
+	block = dasd_alloc_block();
+	if (IS_ERR(block)) {
+		DEV_MESSAGE(KERN_WARNING, device, "%s",
+			    "could not allocate dasd block structure");
+		kfree(device->private);
+		return PTR_ERR(block);
+	}
+	device->block = block;
+	block->base = device;
+
 	/* Read Device Characteristics */
 	rdc_data = (void *) &(private->rdc_data);
 	rdc_data->dev_nr = private->dev_id.devno;
@@ -409,14 +420,14 @@ dasd_diag_check_device(struct dasd_device *device)
 		  sizeof(DASD_DIAG_CMS1)) == 0) {
 		/* get formatted blocksize from label block */
 		bsize = (unsigned int) label->block_size;
-		device->blocks = (unsigned long) label->block_count;
+		block->blocks = (unsigned long) label->block_count;
 	} else
-		device->blocks = end_block;
-	device->bp_block = bsize;
-	device->s2b_shift = 0;	/* bits to shift 512 to get a block */
+		block->blocks = end_block;
+	block->bp_block = bsize;
+	block->s2b_shift = 0;	/* bits to shift 512 to get a block */
 	for (sb = 512; sb < bsize; sb = sb << 1)
-		device->s2b_shift++;
-	rc = mdsk_init_io(device, device->bp_block, 0, NULL);
+		block->s2b_shift++;
+	rc = mdsk_init_io(device, block->bp_block, 0, NULL);
 	if (rc) {
 		DEV_MESSAGE(KERN_WARNING, device, "DIAG initialization "
 			"failed (rc=%d)", rc);
@@ -424,9 +435,9 @@ dasd_diag_check_device(struct dasd_device *device)
 	} else {
 		DEV_MESSAGE(KERN_INFO, device,
 			    "(%ld B/blk): %ldkB",
-			    (unsigned long) device->bp_block,
-			    (unsigned long) (device->blocks <<
-				device->s2b_shift) >> 1);
+			    (unsigned long) block->bp_block,
+			    (unsigned long) (block->blocks <<
+				block->s2b_shift) >> 1);
 	}
 out:
 	free_page((long) label);
@@ -436,22 +447,16 @@ out:
 /* Fill in virtual disk geometry for device. Return zero on success, non-zero
  * otherwise. */
 static int
-dasd_diag_fill_geometry(struct dasd_device *device, struct hd_geometry *geo)
+dasd_diag_fill_geometry(struct dasd_block *block, struct hd_geometry *geo)
 {
-	if (dasd_check_blocksize(device->bp_block) != 0)
+	if (dasd_check_blocksize(block->bp_block) != 0)
 		return -EINVAL;
-	geo->cylinders = (device->blocks << device->s2b_shift) >> 10;
+	geo->cylinders = (block->blocks << block->s2b_shift) >> 10;
 	geo->heads = 16;
-	geo->sectors = 128 >> device->s2b_shift;
+	geo->sectors = 128 >> block->s2b_shift;
 	return 0;
 }
 
-static dasd_era_t
-dasd_diag_examine_error(struct dasd_ccw_req * cqr, struct irb * stat)
-{
-	return dasd_era_fatal;
-}
-
 static dasd_erp_fn_t
 dasd_diag_erp_action(struct dasd_ccw_req * cqr)
 {
@@ -466,8 +471,9 @@ dasd_diag_erp_postaction(struct dasd_ccw_req * cqr)
 
 /* Create DASD request from block device request. Return pointer to new
  * request on success, ERR_PTR otherwise. */
-static struct dasd_ccw_req *
-dasd_diag_build_cp(struct dasd_device * device, struct request *req)
+static struct dasd_ccw_req *dasd_diag_build_cp(struct dasd_device *memdev,
+					       struct dasd_block *block,
+					       struct request *req)
 {
 	struct dasd_ccw_req *cqr;
 	struct dasd_diag_req *dreq;
@@ -486,17 +492,17 @@ dasd_diag_build_cp(struct dasd_device * device, struct request *req)
 		rw_cmd = MDSK_WRITE_REQ;
 	else
 		return ERR_PTR(-EINVAL);
-	blksize = device->bp_block;
+	blksize = block->bp_block;
 	/* Calculate record id of first and last block. */
-	first_rec = req->sector >> device->s2b_shift;
-	last_rec = (req->sector + req->nr_sectors - 1) >> device->s2b_shift;
+	first_rec = req->sector >> block->s2b_shift;
+	last_rec = (req->sector + req->nr_sectors - 1) >> block->s2b_shift;
 	/* Check struct bio and count the number of blocks for the request. */
 	count = 0;
 	rq_for_each_segment(bv, req, iter) {
 		if (bv->bv_len & (blksize - 1))
 			/* Fba can only do full blocks. */
 			return ERR_PTR(-EINVAL);
-		count += bv->bv_len >> (device->s2b_shift + 9);
+		count += bv->bv_len >> (block->s2b_shift + 9);
 	}
 	/* Paranoia. */
 	if (count != last_rec - first_rec + 1)
@@ -505,7 +511,7 @@ dasd_diag_build_cp(struct dasd_device * device, struct request *req)
 	datasize = sizeof(struct dasd_diag_req) +
 		count*sizeof(struct dasd_diag_bio);
 	cqr = dasd_smalloc_request(dasd_diag_discipline.name, 0,
-				   datasize, device);
+				   datasize, memdev);
 	if (IS_ERR(cqr))
 		return cqr;
 
@@ -529,7 +535,9 @@ dasd_diag_build_cp(struct dasd_device * device, struct request *req)
 	cqr->buildclk = get_clock();
 	if (req->cmd_flags & REQ_FAILFAST)
 		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
-	cqr->device = device;
+	cqr->startdev = memdev;
+	cqr->memdev = memdev;
+	cqr->block = block;
 	cqr->expires = DIAG_TIMEOUT;
 	cqr->status = DASD_CQR_FILLED;
 	return cqr;
@@ -543,10 +551,15 @@ dasd_diag_free_cp(struct dasd_ccw_req *cqr, struct request *req)
 	int status;
 
 	status = cqr->status == DASD_CQR_DONE;
-	dasd_sfree_request(cqr, cqr->device);
+	dasd_sfree_request(cqr, cqr->memdev);
 	return status;
 }
 
+static void dasd_diag_handle_terminated_request(struct dasd_ccw_req *cqr)
+{
+	cqr->status = DASD_CQR_FILLED;
+};
+
 /* Fill in IOCTL data for device. */
 static int
 dasd_diag_fill_info(struct dasd_device * device,
@@ -583,7 +596,7 @@ static struct dasd_discipline dasd_diag_discipline = {
 	.fill_geometry = dasd_diag_fill_geometry,
 	.start_IO = dasd_start_diag,
 	.term_IO = dasd_diag_term_IO,
-	.examine_error = dasd_diag_examine_error,
+	.handle_terminated_request = dasd_diag_handle_terminated_request,
 	.erp_action = dasd_diag_erp_action,
 	.erp_postaction = dasd_diag_erp_postaction,
 	.build_cp = dasd_diag_build_cp,