add idl4k kernel firmware version 1.13.0.105

kernel/drivers/s390/Makefile

@@ -0,0 +1,8 @@
+#
+# Makefile for the S/390 specific device drivers
+#
+
+obj-y += cio/ block/ char/ crypto/ net/ scsi/ kvm/
+
+drivers-y += drivers/s390/built-in.o
+

kernel/drivers/s390/block/Kconfig

@@ -0,0 +1,64 @@
+comment "S/390 block device drivers"
+	depends on S390 && BLOCK
+
+config BLK_DEV_XPRAM
+	tristate "XPRAM disk support"
+	depends on S390 && BLOCK
+	help
+	  Select this option if you want to use your expanded storage on S/390
+	  or zSeries as a disk.  This is useful as a _fast_ swap device if you
+	  want to access more than 2G of memory when running in 31 bit mode.
+	  This option is also available as a module which will be called
+	  xpram.  If unsure, say "N".
+
+config DCSSBLK
+	tristate "DCSSBLK support"
+	depends on S390 && BLOCK
+	help
+	  Support for dcss block device
+
+config DASD
+	tristate "Support for DASD devices"
+	depends on CCW && BLOCK
+	select IOSCHED_DEADLINE
+	help
+	  Enable this option if you want to access DASDs directly utilizing
+	  S/390s channel subsystem commands. This is necessary for running
+	  natively on a single image or an LPAR.
+
+config DASD_PROFILE
+	bool "Profiling support for dasd devices"
+	depends on DASD
+	help
+	  Enable this option if you want to see profiling information
+          in /proc/dasd/statistics.
+
+config DASD_ECKD
+	tristate "Support for ECKD Disks"
+	depends on DASD
+	help
+	  ECKD devices are the most commonly used devices. You should enable
+	  this option unless you are very sure to have no ECKD device.
+
+config DASD_FBA
+	tristate "Support for FBA  Disks"
+	depends on DASD
+	help
+	  Select this option to be able to access FBA devices. It is safe to
+	  say "Y".
+
+config DASD_DIAG
+	tristate "Support for DIAG access to Disks"
+	depends on DASD
+	help
+	  Select this option if you want to use Diagnose250 command to access
+	  Disks under VM.  If you are not running under VM or unsure what it is,
+	  say "N".
+
+config DASD_EER
+	bool "Extended error reporting (EER)"
+	depends on DASD
+	help
+	  This driver provides a character device interface to the
+	  DASD extended error reporting. This is only needed if you want to
+	  use applications written for the EER facility.

kernel/drivers/s390/block/Makefile

@@ -0,0 +1,19 @@
+#
+# S/390 block devices
+#
+
+dasd_eckd_mod-objs := dasd_eckd.o dasd_3990_erp.o dasd_alias.o
+dasd_fba_mod-objs  := dasd_fba.o
+dasd_diag_mod-objs := dasd_diag.o
+dasd_mod-objs      := dasd.o dasd_ioctl.o dasd_proc.o dasd_devmap.o \
+			dasd_genhd.o dasd_erp.o
+ifdef CONFIG_DASD_EER
+dasd_mod-objs      += dasd_eer.o
+endif
+
+obj-$(CONFIG_DASD) += dasd_mod.o
+obj-$(CONFIG_DASD_DIAG) += dasd_diag_mod.o
+obj-$(CONFIG_DASD_ECKD) += dasd_eckd_mod.o
+obj-$(CONFIG_DASD_FBA)  += dasd_fba_mod.o
+obj-$(CONFIG_BLK_DEV_XPRAM) += xpram.o
+obj-$(CONFIG_DCSSBLK) += dcssblk.o

kernel/drivers/s390/block/dasd_alias.c

@@ -0,0 +1,910 @@
+/*
+ * PAV alias management for the DASD ECKD discipline
+ *
+ * Copyright IBM Corporation, 2007
+ * Author(s): Stefan Weinhuber <wein@de.ibm.com>
+ */
+
+#define KMSG_COMPONENT "dasd-eckd"
+
+#include <linux/list.h>
+#include <asm/ebcdic.h>
+#include "dasd_int.h"
+#include "dasd_eckd.h"
+
+#ifdef PRINTK_HEADER
+#undef PRINTK_HEADER
+#endif				/* PRINTK_HEADER */
+#define PRINTK_HEADER "dasd(eckd):"
+
+
+/*
+ * General concept of alias management:
+ * - PAV and DASD alias management is specific to the eckd discipline.
+ * - A device is connected to an lcu as long as the device exists.
+ *   dasd_alias_make_device_known_to_lcu will be called wenn the
+ *   device is checked by the eckd discipline and
+ *   dasd_alias_disconnect_device_from_lcu will be called
+ *   before the device is deleted.
+ * - The dasd_alias_add_device / dasd_alias_remove_device
+ *   functions mark the point when a device is 'ready for service'.
+ * - A summary unit check is a rare occasion, but it is mandatory to
+ *   support it. It requires some complex recovery actions before the
+ *   devices can be used again (see dasd_alias_handle_summary_unit_check).
+ * - dasd_alias_get_start_dev will find an alias device that can be used
+ *   instead of the base device and does some (very simple) load balancing.
+ *   This is the function that gets called for each I/O, so when improving
+ *   something, this function should get faster or better, the rest has just
+ *   to be correct.
+ */
+
+
+static void summary_unit_check_handling_work(struct work_struct *);
+static void lcu_update_work(struct work_struct *);
+static int _schedule_lcu_update(struct alias_lcu *, struct dasd_device *);
+
+static struct alias_root aliastree = {
+	.serverlist = LIST_HEAD_INIT(aliastree.serverlist),
+	.lock = __SPIN_LOCK_UNLOCKED(aliastree.lock),
+};
+
+static struct alias_server *_find_server(struct dasd_uid *uid)
+{
+	struct alias_server *pos;
+	list_for_each_entry(pos, &aliastree.serverlist, server) {
+		if (!strncmp(pos->uid.vendor, uid->vendor,
+			     sizeof(uid->vendor))
+		    && !strncmp(pos->uid.serial, uid->serial,
+				sizeof(uid->serial)))
+			return pos;
+	};
+	return NULL;
+}
+
+static struct alias_lcu *_find_lcu(struct alias_server *server,
+				   struct dasd_uid *uid)
+{
+	struct alias_lcu *pos;
+	list_for_each_entry(pos, &server->lculist, lcu) {
+		if (pos->uid.ssid == uid->ssid)
+			return pos;
+	};
+	return NULL;
+}
+
+static struct alias_pav_group *_find_group(struct alias_lcu *lcu,
+					   struct dasd_uid *uid)
+{
+	struct alias_pav_group *pos;
+	__u8 search_unit_addr;
+
+	/* for hyper pav there is only one group */
+	if (lcu->pav == HYPER_PAV) {
+		if (list_empty(&lcu->grouplist))
+			return NULL;
+		else
+			return list_first_entry(&lcu->grouplist,
+						struct alias_pav_group, group);
+	}
+
+	/* for base pav we have to find the group that matches the base */
+	if (uid->type == UA_BASE_DEVICE)
+		search_unit_addr = uid->real_unit_addr;
+	else
+		search_unit_addr = uid->base_unit_addr;
+	list_for_each_entry(pos, &lcu->grouplist, group) {
+		if (pos->uid.base_unit_addr == search_unit_addr &&
+		    !strncmp(pos->uid.vduit, uid->vduit, sizeof(uid->vduit)))
+			return pos;
+	};
+	return NULL;
+}
+
+static struct alias_server *_allocate_server(struct dasd_uid *uid)
+{
+	struct alias_server *server;
+
+	server = kzalloc(sizeof(*server), GFP_KERNEL);
+	if (!server)
+		return ERR_PTR(-ENOMEM);
+	memcpy(server->uid.vendor, uid->vendor, sizeof(uid->vendor));
+	memcpy(server->uid.serial, uid->serial, sizeof(uid->serial));
+	INIT_LIST_HEAD(&server->server);
+	INIT_LIST_HEAD(&server->lculist);
+	return server;
+}
+
+static void _free_server(struct alias_server *server)
+{
+	kfree(server);
+}
+
+static struct alias_lcu *_allocate_lcu(struct dasd_uid *uid)
+{
+	struct alias_lcu *lcu;
+
+	lcu = kzalloc(sizeof(*lcu), GFP_KERNEL);
+	if (!lcu)
+		return ERR_PTR(-ENOMEM);
+	lcu->uac = kzalloc(sizeof(*(lcu->uac)), GFP_KERNEL | GFP_DMA);
+	if (!lcu->uac)
+		goto out_err1;
+	lcu->rsu_cqr = kzalloc(sizeof(*lcu->rsu_cqr), GFP_KERNEL | GFP_DMA);
+	if (!lcu->rsu_cqr)
+		goto out_err2;
+	lcu->rsu_cqr->cpaddr = kzalloc(sizeof(struct ccw1),
+				       GFP_KERNEL | GFP_DMA);
+	if (!lcu->rsu_cqr->cpaddr)
+		goto out_err3;
+	lcu->rsu_cqr->data = kzalloc(16, GFP_KERNEL | GFP_DMA);
+	if (!lcu->rsu_cqr->data)
+		goto out_err4;
+
+	memcpy(lcu->uid.vendor, uid->vendor, sizeof(uid->vendor));
+	memcpy(lcu->uid.serial, uid->serial, sizeof(uid->serial));
+	lcu->uid.ssid = uid->ssid;
+	lcu->pav = NO_PAV;
+	lcu->flags = NEED_UAC_UPDATE | UPDATE_PENDING;
+	INIT_LIST_HEAD(&lcu->lcu);
+	INIT_LIST_HEAD(&lcu->inactive_devices);
+	INIT_LIST_HEAD(&lcu->active_devices);
+	INIT_LIST_HEAD(&lcu->grouplist);
+	INIT_WORK(&lcu->suc_data.worker, summary_unit_check_handling_work);
+	INIT_DELAYED_WORK(&lcu->ruac_data.dwork, lcu_update_work);
+	spin_lock_init(&lcu->lock);
+	return lcu;
+
+out_err4:
+	kfree(lcu->rsu_cqr->cpaddr);
+out_err3:
+	kfree(lcu->rsu_cqr);
+out_err2:
+	kfree(lcu->uac);
+out_err1:
+	kfree(lcu);
+	return ERR_PTR(-ENOMEM);
+}
+
+static void _free_lcu(struct alias_lcu *lcu)
+{
+	kfree(lcu->rsu_cqr->data);
+	kfree(lcu->rsu_cqr->cpaddr);
+	kfree(lcu->rsu_cqr);
+	kfree(lcu->uac);
+	kfree(lcu);
+}
+
+/*
+ * This is the function that will allocate all the server and lcu data,
+ * so this function must be called first for a new device.
+ * If the return value is 1, the lcu was already known before, if it
+ * is 0, this is a new lcu.
+ * Negative return code indicates that something went wrong (e.g. -ENOMEM)
+ */
+int dasd_alias_make_device_known_to_lcu(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private;
+	unsigned long flags;
+	struct alias_server *server, *newserver;
+	struct alias_lcu *lcu, *newlcu;
+	int is_lcu_known;
+	struct dasd_uid *uid;
+
+	private = (struct dasd_eckd_private *) device->private;
+	uid = &private->uid;
+	spin_lock_irqsave(&aliastree.lock, flags);
+	is_lcu_known = 1;
+	server = _find_server(uid);
+	if (!server) {
+		spin_unlock_irqrestore(&aliastree.lock, flags);
+		newserver = _allocate_server(uid);
+		if (IS_ERR(newserver))
+			return PTR_ERR(newserver);
+		spin_lock_irqsave(&aliastree.lock, flags);
+		server = _find_server(uid);
+		if (!server) {
+			list_add(&newserver->server, &aliastree.serverlist);
+			server = newserver;
+			is_lcu_known = 0;
+		} else {
+			/* someone was faster */
+			_free_server(newserver);
+		}
+	}
+
+	lcu = _find_lcu(server, uid);
+	if (!lcu) {
+		spin_unlock_irqrestore(&aliastree.lock, flags);
+		newlcu = _allocate_lcu(uid);
+		if (IS_ERR(newlcu))
+			return PTR_ERR(lcu);
+		spin_lock_irqsave(&aliastree.lock, flags);
+		lcu = _find_lcu(server, uid);
+		if (!lcu) {
+			list_add(&newlcu->lcu, &server->lculist);
+			lcu = newlcu;
+			is_lcu_known = 0;
+		} else {
+			/* someone was faster */
+			_free_lcu(newlcu);
+		}
+		is_lcu_known = 0;
+	}
+	spin_lock(&lcu->lock);
+	list_add(&device->alias_list, &lcu->inactive_devices);
+	private->lcu = lcu;
+	spin_unlock(&lcu->lock);
+	spin_unlock_irqrestore(&aliastree.lock, flags);
+
+	return is_lcu_known;
+}
+
+/*
+ * This function removes a device from the scope of alias management.
+ * The complicated part is to make sure that it is not in use by
+ * any of the workers. If necessary cancel the work.
+ */
+void dasd_alias_disconnect_device_from_lcu(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private;
+	unsigned long flags;
+	struct alias_lcu *lcu;
+	struct alias_server *server;
+	int was_pending;
+
+	private = (struct dasd_eckd_private *) device->private;
+	lcu = private->lcu;
+	spin_lock_irqsave(&lcu->lock, flags);
+	list_del_init(&device->alias_list);
+	/* make sure that the workers don't use this device */
+	if (device == lcu->suc_data.device) {
+		spin_unlock_irqrestore(&lcu->lock, flags);
+		cancel_work_sync(&lcu->suc_data.worker);
+		spin_lock_irqsave(&lcu->lock, flags);
+		if (device == lcu->suc_data.device)
+			lcu->suc_data.device = NULL;
+	}
+	was_pending = 0;
+	if (device == lcu->ruac_data.device) {
+		spin_unlock_irqrestore(&lcu->lock, flags);
+		was_pending = 1;
+		cancel_delayed_work_sync(&lcu->ruac_data.dwork);
+		spin_lock_irqsave(&lcu->lock, flags);
+		if (device == lcu->ruac_data.device)
+			lcu->ruac_data.device = NULL;
+	}
+	private->lcu = NULL;
+	spin_unlock_irqrestore(&lcu->lock, flags);
+
+	spin_lock_irqsave(&aliastree.lock, flags);
+	spin_lock(&lcu->lock);
+	if (list_empty(&lcu->grouplist) &&
+	    list_empty(&lcu->active_devices) &&
+	    list_empty(&lcu->inactive_devices)) {
+		list_del(&lcu->lcu);
+		spin_unlock(&lcu->lock);
+		_free_lcu(lcu);
+		lcu = NULL;
+	} else {
+		if (was_pending)
+			_schedule_lcu_update(lcu, NULL);
+		spin_unlock(&lcu->lock);
+	}
+	server = _find_server(&private->uid);
+	if (server && list_empty(&server->lculist)) {
+		list_del(&server->server);
+		_free_server(server);
+	}
+	spin_unlock_irqrestore(&aliastree.lock, flags);
+}
+
+/*
+ * This function assumes that the unit address configuration stored
+ * in the lcu is up to date and will update the device uid before
+ * adding it to a pav group.
+ */
+static int _add_device_to_lcu(struct alias_lcu *lcu,
+			      struct dasd_device *device)
+{
+
+	struct dasd_eckd_private *private;
+	struct alias_pav_group *group;
+	struct dasd_uid *uid;
+
+	private = (struct dasd_eckd_private *) device->private;
+	uid = &private->uid;
+	uid->type = lcu->uac->unit[uid->real_unit_addr].ua_type;
+	uid->base_unit_addr = lcu->uac->unit[uid->real_unit_addr].base_ua;
+	dasd_set_uid(device->cdev, &private->uid);
+
+	/* if we have no PAV anyway, we don't need to bother with PAV groups */
+	if (lcu->pav == NO_PAV) {
+		list_move(&device->alias_list, &lcu->active_devices);
+		return 0;
+	}
+
+	group = _find_group(lcu, uid);
+	if (!group) {
+		group = kzalloc(sizeof(*group), GFP_ATOMIC);
+		if (!group)
+			return -ENOMEM;
+		memcpy(group->uid.vendor, uid->vendor, sizeof(uid->vendor));
+		memcpy(group->uid.serial, uid->serial, sizeof(uid->serial));
+		group->uid.ssid = uid->ssid;
+		if (uid->type == UA_BASE_DEVICE)
+			group->uid.base_unit_addr = uid->real_unit_addr;
+		else
+			group->uid.base_unit_addr = uid->base_unit_addr;
+		memcpy(group->uid.vduit, uid->vduit, sizeof(uid->vduit));
+		INIT_LIST_HEAD(&group->group);
+		INIT_LIST_HEAD(&group->baselist);
+		INIT_LIST_HEAD(&group->aliaslist);
+		list_add(&group->group, &lcu->grouplist);
+	}
+	if (uid->type == UA_BASE_DEVICE)
+		list_move(&device->alias_list, &group->baselist);
+	else
+		list_move(&device->alias_list, &group->aliaslist);
+	private->pavgroup = group;
+	return 0;
+};
+
+static void _remove_device_from_lcu(struct alias_lcu *lcu,
+				    struct dasd_device *device)
+{
+	struct dasd_eckd_private *private;
+	struct alias_pav_group *group;
+
+	private = (struct dasd_eckd_private *) device->private;
+	list_move(&device->alias_list, &lcu->inactive_devices);
+	group = private->pavgroup;
+	if (!group)
+		return;
+	private->pavgroup = NULL;
+	if (list_empty(&group->baselist) && list_empty(&group->aliaslist)) {
+		list_del(&group->group);
+		kfree(group);
+		return;
+	}
+	if (group->next == device)
+		group->next = NULL;
+};
+
+static int read_unit_address_configuration(struct dasd_device *device,
+					   struct alias_lcu *lcu)
+{
+	struct dasd_psf_prssd_data *prssdp;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	int rc;
+	unsigned long flags;
+
+	cqr = dasd_kmalloc_request(DASD_ECKD_MAGIC, 1 /* PSF */	+ 1 /* RSSD */,
+				   (sizeof(struct dasd_psf_prssd_data)),
+				   device);
+	if (IS_ERR(cqr))
+		return PTR_ERR(cqr);
+	cqr->startdev = device;
+	cqr->memdev = device;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	cqr->retries = 10;
+	cqr->expires = 20 * HZ;
+
+	/* Prepare for Read Subsystem Data */
+	prssdp = (struct dasd_psf_prssd_data *) cqr->data;
+	memset(prssdp, 0, sizeof(struct dasd_psf_prssd_data));
+	prssdp->order = PSF_ORDER_PRSSD;
+	prssdp->suborder = 0x0e;	/* Read unit address configuration */
+	/* all other bytes of prssdp must be zero */
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_PSF;
+	ccw->count = sizeof(struct dasd_psf_prssd_data);
+	ccw->flags |= CCW_FLAG_CC;
+	ccw->cda = (__u32)(addr_t) prssdp;
+
+	/* Read Subsystem Data - feature codes */
+	memset(lcu->uac, 0, sizeof(*(lcu->uac)));
+
+	ccw++;
+	ccw->cmd_code = DASD_ECKD_CCW_RSSD;
+	ccw->count = sizeof(*(lcu->uac));
+	ccw->cda = (__u32)(addr_t) lcu->uac;
+
+	cqr->buildclk = get_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	/* need to unset flag here to detect race with summary unit check */
+	spin_lock_irqsave(&lcu->lock, flags);
+	lcu->flags &= ~NEED_UAC_UPDATE;
+	spin_unlock_irqrestore(&lcu->lock, flags);
+
+	do {
+		rc = dasd_sleep_on(cqr);
+	} while (rc && (cqr->retries > 0));
+	if (rc) {
+		spin_lock_irqsave(&lcu->lock, flags);
+		lcu->flags |= NEED_UAC_UPDATE;
+		spin_unlock_irqrestore(&lcu->lock, flags);
+	}
+	dasd_kfree_request(cqr, cqr->memdev);
+	return rc;
+}
+
+static int _lcu_update(struct dasd_device *refdev, struct alias_lcu *lcu)
+{
+	unsigned long flags;
+	struct alias_pav_group *pavgroup, *tempgroup;
+	struct dasd_device *device, *tempdev;
+	int i, rc;
+	struct dasd_eckd_private *private;
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	list_for_each_entry_safe(pavgroup, tempgroup, &lcu->grouplist, group) {
+		list_for_each_entry_safe(device, tempdev, &pavgroup->baselist,
+					 alias_list) {
+			list_move(&device->alias_list, &lcu->active_devices);
+			private = (struct dasd_eckd_private *) device->private;
+			private->pavgroup = NULL;
+		}
+		list_for_each_entry_safe(device, tempdev, &pavgroup->aliaslist,
+					 alias_list) {
+			list_move(&device->alias_list, &lcu->active_devices);
+			private = (struct dasd_eckd_private *) device->private;
+			private->pavgroup = NULL;
+		}
+		list_del(&pavgroup->group);
+		kfree(pavgroup);
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+
+	rc = read_unit_address_configuration(refdev, lcu);
+	if (rc)
+		return rc;
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	lcu->pav = NO_PAV;
+	for (i = 0; i < MAX_DEVICES_PER_LCU; ++i) {
+		switch (lcu->uac->unit[i].ua_type) {
+		case UA_BASE_PAV_ALIAS:
+			lcu->pav = BASE_PAV;
+			break;
+		case UA_HYPER_PAV_ALIAS:
+			lcu->pav = HYPER_PAV;
+			break;
+		}
+		if (lcu->pav != NO_PAV)
+			break;
+	}
+
+	list_for_each_entry_safe(device, tempdev, &lcu->active_devices,
+				 alias_list) {
+		_add_device_to_lcu(lcu, device);
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+	return 0;
+}
+
+static void lcu_update_work(struct work_struct *work)
+{
+	struct alias_lcu *lcu;
+	struct read_uac_work_data *ruac_data;
+	struct dasd_device *device;
+	unsigned long flags;
+	int rc;
+
+	ruac_data = container_of(work, struct read_uac_work_data, dwork.work);
+	lcu = container_of(ruac_data, struct alias_lcu, ruac_data);
+	device = ruac_data->device;
+	rc = _lcu_update(device, lcu);
+	/*
+	 * Need to check flags again, as there could have been another
+	 * prepare_update or a new device a new device while we were still
+	 * processing the data
+	 */
+	spin_lock_irqsave(&lcu->lock, flags);
+	if (rc || (lcu->flags & NEED_UAC_UPDATE)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "could not update"
+			    " alias data in lcu (rc = %d), retry later", rc);
+		schedule_delayed_work(&lcu->ruac_data.dwork, 30*HZ);
+	} else {
+		lcu->ruac_data.device = NULL;
+		lcu->flags &= ~UPDATE_PENDING;
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+}
+
+static int _schedule_lcu_update(struct alias_lcu *lcu,
+				struct dasd_device *device)
+{
+	struct dasd_device *usedev = NULL;
+	struct alias_pav_group *group;
+
+	lcu->flags |= NEED_UAC_UPDATE;
+	if (lcu->ruac_data.device) {
+		/* already scheduled or running */
+		return 0;
+	}
+	if (device && !list_empty(&device->alias_list))
+		usedev = device;
+
+	if (!usedev && !list_empty(&lcu->grouplist)) {
+		group = list_first_entry(&lcu->grouplist,
+					 struct alias_pav_group, group);
+		if (!list_empty(&group->baselist))
+			usedev = list_first_entry(&group->baselist,
+						  struct dasd_device,
+						  alias_list);
+		else if (!list_empty(&group->aliaslist))
+			usedev = list_first_entry(&group->aliaslist,
+						  struct dasd_device,
+						  alias_list);
+	}
+	if (!usedev && !list_empty(&lcu->active_devices)) {
+		usedev = list_first_entry(&lcu->active_devices,
+					  struct dasd_device, alias_list);
+	}
+	/*
+	 * if we haven't found a proper device yet, give up for now, the next
+	 * device that will be set active will trigger an lcu update
+	 */
+	if (!usedev)
+		return -EINVAL;
+	lcu->ruac_data.device = usedev;
+	schedule_delayed_work(&lcu->ruac_data.dwork, 0);
+	return 0;
+}
+
+int dasd_alias_add_device(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private;
+	struct alias_lcu *lcu;
+	unsigned long flags;
+	int rc;
+
+	private = (struct dasd_eckd_private *) device->private;
+	lcu = private->lcu;
+	rc = 0;
+	spin_lock_irqsave(&lcu->lock, flags);
+	if (!(lcu->flags & UPDATE_PENDING)) {
+		rc = _add_device_to_lcu(lcu, device);
+		if (rc)
+			lcu->flags |= UPDATE_PENDING;
+	}
+	if (lcu->flags & UPDATE_PENDING) {
+		list_move(&device->alias_list, &lcu->active_devices);
+		_schedule_lcu_update(lcu, device);
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+	return rc;
+}
+
+int dasd_alias_remove_device(struct dasd_device *device)
+{
+	struct dasd_eckd_private *private;
+	struct alias_lcu *lcu;
+	unsigned long flags;
+
+	private = (struct dasd_eckd_private *) device->private;
+	lcu = private->lcu;
+	spin_lock_irqsave(&lcu->lock, flags);
+	_remove_device_from_lcu(lcu, device);
+	spin_unlock_irqrestore(&lcu->lock, flags);
+	return 0;
+}
+
+struct dasd_device *dasd_alias_get_start_dev(struct dasd_device *base_device)
+{
+
+	struct dasd_device *alias_device;
+	struct alias_pav_group *group;
+	struct alias_lcu *lcu;
+	struct dasd_eckd_private *private, *alias_priv;
+	unsigned long flags;
+
+	private = (struct dasd_eckd_private *) base_device->private;
+	group = private->pavgroup;
+	lcu = private->lcu;
+	if (!group || !lcu)
+		return NULL;
+	if (lcu->pav == NO_PAV ||
+	    lcu->flags & (NEED_UAC_UPDATE | UPDATE_PENDING))
+		return NULL;
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	alias_device = group->next;
+	if (!alias_device) {
+		if (list_empty(&group->aliaslist)) {
+			spin_unlock_irqrestore(&lcu->lock, flags);
+			return NULL;
+		} else {
+			alias_device = list_first_entry(&group->aliaslist,
+							struct dasd_device,
+							alias_list);
+		}
+	}
+	if (list_is_last(&alias_device->alias_list, &group->aliaslist))
+		group->next = list_first_entry(&group->aliaslist,
+					       struct dasd_device, alias_list);
+	else
+		group->next = list_first_entry(&alias_device->alias_list,
+					       struct dasd_device, alias_list);
+	spin_unlock_irqrestore(&lcu->lock, flags);
+	alias_priv = (struct dasd_eckd_private *) alias_device->private;
+	if ((alias_priv->count < private->count) && !alias_device->stopped)
+		return alias_device;
+	else
+		return NULL;
+}
+
+/*
+ * Summary unit check handling depends on the way alias devices
+ * are handled so it is done here rather then in dasd_eckd.c
+ */
+static int reset_summary_unit_check(struct alias_lcu *lcu,
+				    struct dasd_device *device,
+				    char reason)
+{
+	struct dasd_ccw_req *cqr;
+	int rc = 0;
+	struct ccw1 *ccw;
+
+	cqr = lcu->rsu_cqr;
+	strncpy((char *) &cqr->magic, "ECKD", 4);
+	ASCEBC((char *) &cqr->magic, 4);
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_RSCK;
+	ccw->flags = 0 ;
+	ccw->count = 16;
+	ccw->cda = (__u32)(addr_t) cqr->data;
+	((char *)cqr->data)[0] = reason;
+
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	cqr->retries = 255;	/* set retry counter to enable basic ERP */
+	cqr->startdev = device;
+	cqr->memdev = device;
+	cqr->block = NULL;
+	cqr->expires = 5 * HZ;
+	cqr->buildclk = get_clock();
+	cqr->status = DASD_CQR_FILLED;
+
+	rc = dasd_sleep_on_immediatly(cqr);
+	return rc;
+}
+
+static void _restart_all_base_devices_on_lcu(struct alias_lcu *lcu)
+{
+	struct alias_pav_group *pavgroup;
+	struct dasd_device *device;
+	struct dasd_eckd_private *private;
+
+	/* active and inactive list can contain alias as well as base devices */
+	list_for_each_entry(device, &lcu->active_devices, alias_list) {
+		private = (struct dasd_eckd_private *) device->private;
+		if (private->uid.type != UA_BASE_DEVICE)
+			continue;
+		dasd_schedule_block_bh(device->block);
+		dasd_schedule_device_bh(device);
+	}
+	list_for_each_entry(device, &lcu->inactive_devices, alias_list) {
+		private = (struct dasd_eckd_private *) device->private;
+		if (private->uid.type != UA_BASE_DEVICE)
+			continue;
+		dasd_schedule_block_bh(device->block);
+		dasd_schedule_device_bh(device);
+	}
+	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
+		list_for_each_entry(device, &pavgroup->baselist, alias_list) {
+			dasd_schedule_block_bh(device->block);
+			dasd_schedule_device_bh(device);
+		}
+	}
+}
+
+static void flush_all_alias_devices_on_lcu(struct alias_lcu *lcu)
+{
+	struct alias_pav_group *pavgroup;
+	struct dasd_device *device, *temp;
+	struct dasd_eckd_private *private;
+	int rc;
+	unsigned long flags;
+	LIST_HEAD(active);
+
+	/*
+	 * Problem here ist that dasd_flush_device_queue may wait
+	 * for termination of a request to complete. We can't keep
+	 * the lcu lock during that time, so we must assume that
+	 * the lists may have changed.
+	 * Idea: first gather all active alias devices in a separate list,
+	 * then flush the first element of this list unlocked, and afterwards
+	 * check if it is still on the list before moving it to the
+	 * active_devices list.
+	 */
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	list_for_each_entry_safe(device, temp, &lcu->active_devices,
+				 alias_list) {
+		private = (struct dasd_eckd_private *) device->private;
+		if (private->uid.type == UA_BASE_DEVICE)
+			continue;
+		list_move(&device->alias_list, &active);
+	}
+
+	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
+		list_splice_init(&pavgroup->aliaslist, &active);
+	}
+	while (!list_empty(&active)) {
+		device = list_first_entry(&active, struct dasd_device,
+					  alias_list);
+		spin_unlock_irqrestore(&lcu->lock, flags);
+		rc = dasd_flush_device_queue(device);
+		spin_lock_irqsave(&lcu->lock, flags);
+		/*
+		 * only move device around if it wasn't moved away while we
+		 * were waiting for the flush
+		 */
+		if (device == list_first_entry(&active,
+					       struct dasd_device, alias_list))
+			list_move(&device->alias_list, &lcu->active_devices);
+	}
+	spin_unlock_irqrestore(&lcu->lock, flags);
+}
+
+static void __stop_device_on_lcu(struct dasd_device *device,
+				 struct dasd_device *pos)
+{
+	/* If pos == device then device is already locked! */
+	if (pos == device) {
+		pos->stopped |= DASD_STOPPED_SU;
+		return;
+	}
+	spin_lock(get_ccwdev_lock(pos->cdev));
+	pos->stopped |= DASD_STOPPED_SU;
+	spin_unlock(get_ccwdev_lock(pos->cdev));
+}
+
+/*
+ * This function is called in interrupt context, so the
+ * cdev lock for device is already locked!
+ */
+static void _stop_all_devices_on_lcu(struct alias_lcu *lcu,
+				     struct dasd_device *device)
+{
+	struct alias_pav_group *pavgroup;
+	struct dasd_device *pos;
+
+	list_for_each_entry(pos, &lcu->active_devices, alias_list)
+		__stop_device_on_lcu(device, pos);
+	list_for_each_entry(pos, &lcu->inactive_devices, alias_list)
+		__stop_device_on_lcu(device, pos);
+	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
+		list_for_each_entry(pos, &pavgroup->baselist, alias_list)
+			__stop_device_on_lcu(device, pos);
+		list_for_each_entry(pos, &pavgroup->aliaslist, alias_list)
+			__stop_device_on_lcu(device, pos);
+	}
+}
+
+static void _unstop_all_devices_on_lcu(struct alias_lcu *lcu)
+{
+	struct alias_pav_group *pavgroup;
+	struct dasd_device *device;
+	unsigned long flags;
+
+	list_for_each_entry(device, &lcu->active_devices, alias_list) {
+		spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+		device->stopped &= ~DASD_STOPPED_SU;
+		spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	}
+
+	list_for_each_entry(device, &lcu->inactive_devices, alias_list) {
+		spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+		device->stopped &= ~DASD_STOPPED_SU;
+		spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	}
+
+	list_for_each_entry(pavgroup, &lcu->grouplist, group) {
+		list_for_each_entry(device, &pavgroup->baselist, alias_list) {
+			spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+			device->stopped &= ~DASD_STOPPED_SU;
+			spin_unlock_irqrestore(get_ccwdev_lock(device->cdev),
+					       flags);
+		}
+		list_for_each_entry(device, &pavgroup->aliaslist, alias_list) {
+			spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+			device->stopped &= ~DASD_STOPPED_SU;
+			spin_unlock_irqrestore(get_ccwdev_lock(device->cdev),
+					       flags);
+		}
+	}
+}
+
+static void summary_unit_check_handling_work(struct work_struct *work)
+{
+	struct alias_lcu *lcu;
+	struct summary_unit_check_work_data *suc_data;
+	unsigned long flags;
+	struct dasd_device *device;
+
+	suc_data = container_of(work, struct summary_unit_check_work_data,
+				worker);
+	lcu = container_of(suc_data, struct alias_lcu, suc_data);
+	device = suc_data->device;
+
+	/* 1. flush alias devices */
+	flush_all_alias_devices_on_lcu(lcu);
+
+	/* 2. reset summary unit check */
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	device->stopped &= ~(DASD_STOPPED_SU | DASD_STOPPED_PENDING);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	reset_summary_unit_check(lcu, device, suc_data->reason);
+
+	spin_lock_irqsave(&lcu->lock, flags);
+	_unstop_all_devices_on_lcu(lcu);
+	_restart_all_base_devices_on_lcu(lcu);
+	/* 3. read new alias configuration */
+	_schedule_lcu_update(lcu, device);
+	lcu->suc_data.device = NULL;
+	spin_unlock_irqrestore(&lcu->lock, flags);
+}
+
+/*
+ * note: this will be called from int handler context (cdev locked)
+ */
+void dasd_alias_handle_summary_unit_check(struct dasd_device *device,
+					  struct irb *irb)
+{
+	struct alias_lcu *lcu;
+	char reason;
+	struct dasd_eckd_private *private;
+	char *sense;
+
+	private = (struct dasd_eckd_private *) device->private;
+
+	sense = dasd_get_sense(irb);
+	if (sense) {
+		reason = sense[8];
+		DBF_DEV_EVENT(DBF_NOTICE, device, "%s %x",
+			    "eckd handle summary unit check: reason", reason);
+	} else {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "eckd handle summary unit check:"
+			    " no reason code available");
+		return;
+	}
+
+	lcu = private->lcu;
+	if (!lcu) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "device not ready to handle summary"
+			    " unit check (no lcu structure)");
+		return;
+	}
+	spin_lock(&lcu->lock);
+	_stop_all_devices_on_lcu(lcu, device);
+	/* prepare for lcu_update */
+	private->lcu->flags |= NEED_UAC_UPDATE | UPDATE_PENDING;
+	/* If this device is about to be removed just return and wait for
+	 * the next interrupt on a different device
+	 */
+	if (list_empty(&device->alias_list)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "device is in offline processing,"
+			    " don't do summary unit check handling");
+		spin_unlock(&lcu->lock);
+		return;
+	}
+	if (lcu->suc_data.device) {
+		/* already scheduled or running */
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "previous instance of summary unit check worker"
+			    " still pending");
+		spin_unlock(&lcu->lock);
+		return ;
+	}
+	lcu->suc_data.reason = reason;
+	lcu->suc_data.device = device;
+	spin_unlock(&lcu->lock);
+	schedule_work(&lcu->suc_data.worker);
+};

kernel/drivers/s390/block/dasd_diag.c

@@ -0,0 +1,659 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_diag.c
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Based on.......: linux/drivers/s390/block/mdisk.c
+ * ...............: by Hartmunt Penner <hpenner@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000
+ *
+ */
+
+#define KMSG_COMPONENT "dasd-diag"
+
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/hdreg.h>
+#include <linux/bio.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/jiffies.h>
+
+#include <asm/dasd.h>
+#include <asm/debug.h>
+#include <asm/ebcdic.h>
+#include <asm/io.h>
+#include <asm/s390_ext.h>
+#include <asm/todclk.h>
+#include <asm/vtoc.h>
+#include <asm/diag.h>
+
+#include "dasd_int.h"
+#include "dasd_diag.h"
+
+#define PRINTK_HEADER "dasd(diag):"
+
+MODULE_LICENSE("GPL");
+
+/* The maximum number of blocks per request (max_blocks) is dependent on the
+ * amount of storage that is available in the static I/O buffer for each
+ * device. Currently each device gets 2 pages. We want to fit two requests
+ * into the available memory so that we can immediately start the next if one
+ * finishes. */
+#define DIAG_MAX_BLOCKS	(((2 * PAGE_SIZE - sizeof(struct dasd_ccw_req) - \
+			   sizeof(struct dasd_diag_req)) / \
+		           sizeof(struct dasd_diag_bio)) / 2)
+#define DIAG_MAX_RETRIES	32
+#define DIAG_TIMEOUT		50 * HZ
+
+static struct dasd_discipline dasd_diag_discipline;
+
+struct dasd_diag_private {
+	struct dasd_diag_characteristics rdc_data;
+	struct dasd_diag_rw_io iob;
+	struct dasd_diag_init_io iib;
+	blocknum_t pt_block;
+	struct ccw_dev_id dev_id;
+};
+
+struct dasd_diag_req {
+	unsigned int block_count;
+	struct dasd_diag_bio bio[0];
+};
+
+static const u8 DASD_DIAG_CMS1[] = { 0xc3, 0xd4, 0xe2, 0xf1 };/* EBCDIC CMS1 */
+
+/* Perform DIAG250 call with block I/O parameter list iob (input and output)
+ * and function code cmd.
+ * In case of an exception return 3. Otherwise return result of bitwise OR of
+ * resulting condition code and DIAG return code. */
+static inline int dia250(void *iob, int cmd)
+{
+	register unsigned long reg2 asm ("2") = (unsigned long) iob;
+	typedef union {
+		struct dasd_diag_init_io init_io;
+		struct dasd_diag_rw_io rw_io;
+	} addr_type;
+	int rc;
+
+	rc = 3;
+	asm volatile(
+		"	diag	2,%2,0x250\n"
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
+		"	or	%0,3\n"
+		"1:\n"
+		EX_TABLE(0b,1b)
+		: "+d" (rc), "=m" (*(addr_type *) iob)
+		: "d" (cmd), "d" (reg2), "m" (*(addr_type *) iob)
+		: "3", "cc");
+	return rc;
+}
+
+/* Initialize block I/O to DIAG device using the specified blocksize and
+ * block offset. On success, return zero and set end_block to contain the
+ * number of blocks on the device minus the specified offset. Return non-zero
+ * otherwise. */
+static inline int
+mdsk_init_io(struct dasd_device *device, unsigned int blocksize,
+	     blocknum_t offset, blocknum_t *end_block)
+{
+	struct dasd_diag_private *private;
+	struct dasd_diag_init_io *iib;
+	int rc;
+
+	private = (struct dasd_diag_private *) device->private;
+	iib = &private->iib;
+	memset(iib, 0, sizeof (struct dasd_diag_init_io));
+
+	iib->dev_nr = private->dev_id.devno;
+	iib->block_size = blocksize;
+	iib->offset = offset;
+	iib->flaga = DASD_DIAG_FLAGA_DEFAULT;
+
+	rc = dia250(iib, INIT_BIO);
+
+	if ((rc & 3) == 0 && end_block)
+		*end_block = iib->end_block;
+
+	return rc;
+}
+
+/* Remove block I/O environment for device. Return zero on success, non-zero
+ * otherwise. */
+static inline int
+mdsk_term_io(struct dasd_device * device)
+{
+	struct dasd_diag_private *private;
+	struct dasd_diag_init_io *iib;
+	int rc;
+
+	private = (struct dasd_diag_private *) device->private;
+	iib = &private->iib;
+	memset(iib, 0, sizeof (struct dasd_diag_init_io));
+	iib->dev_nr = private->dev_id.devno;
+	rc = dia250(iib, TERM_BIO);
+	return rc;
+}
+
+/* Error recovery for failed DIAG requests - try to reestablish the DIAG
+ * environment. */
+static void
+dasd_diag_erp(struct dasd_device *device)
+{
+	int rc;
+
+	mdsk_term_io(device);
+	rc = mdsk_init_io(device, device->block->bp_block, 0, NULL);
+	if (rc == 4) {
+		if (!(device->features & DASD_FEATURE_READONLY)) {
+			dev_warn(&device->cdev->dev,
+				 "The access mode of a DIAG device changed"
+				 " to read-only");
+			device->features |= DASD_FEATURE_READONLY;
+		}
+		rc = 0;
+	}
+	if (rc)
+		dev_warn(&device->cdev->dev, "DIAG ERP failed with "
+			    "rc=%d\n", rc);
+}
+
+/* Start a given request at the device. Return zero on success, non-zero
+ * otherwise. */
+static int
+dasd_start_diag(struct dasd_ccw_req * cqr)
+{
+	struct dasd_device *device;
+	struct dasd_diag_private *private;
+	struct dasd_diag_req *dreq;
+	int rc;
+
+	device = cqr->startdev;
+	if (cqr->retries < 0) {
+		DBF_DEV_EVENT(DBF_ERR, device, "DIAG start_IO: request %p "
+			    "- no retry left)", cqr);
+		cqr->status = DASD_CQR_ERROR;
+		return -EIO;
+	}
+	private = (struct dasd_diag_private *) device->private;
+	dreq = (struct dasd_diag_req *) cqr->data;
+
+	private->iob.dev_nr = private->dev_id.devno;
+	private->iob.key = 0;
+	private->iob.flags = DASD_DIAG_RWFLAG_ASYNC;
+	private->iob.block_count = dreq->block_count;
+	private->iob.interrupt_params = (addr_t) cqr;
+	private->iob.bio_list = dreq->bio;
+	private->iob.flaga = DASD_DIAG_FLAGA_DEFAULT;
+
+	cqr->startclk = get_clock();
+	cqr->starttime = jiffies;
+	cqr->retries--;
+
+	rc = dia250(&private->iob, RW_BIO);
+	switch (rc) {
+	case 0: /* Synchronous I/O finished successfully */
+		cqr->stopclk = get_clock();
+		cqr->status = DASD_CQR_SUCCESS;
+		/* Indicate to calling function that only a dasd_schedule_bh()
+		   and no timer is needed */
+                rc = -EACCES;
+		break;
+	case 8: /* Asynchronous I/O was started */
+		cqr->status = DASD_CQR_IN_IO;
+		rc = 0;
+		break;
+	default: /* Error condition */
+		cqr->status = DASD_CQR_QUEUED;
+		DBF_DEV_EVENT(DBF_WARNING, device, "dia250 returned rc=%d", rc);
+		dasd_diag_erp(device);
+		rc = -EIO;
+		break;
+	}
+	cqr->intrc = rc;
+	return rc;
+}
+
+/* Terminate given request at the device. */
+static int
+dasd_diag_term_IO(struct dasd_ccw_req * cqr)
+{
+	struct dasd_device *device;
+
+	device = cqr->startdev;
+	mdsk_term_io(device);
+	mdsk_init_io(device, device->block->bp_block, 0, NULL);
+	cqr->status = DASD_CQR_CLEAR_PENDING;
+	cqr->stopclk = get_clock();
+	dasd_schedule_device_bh(device);
+	return 0;
+}
+
+/* Handle external interruption. */
+static void
+dasd_ext_handler(__u16 code)
+{
+	struct dasd_ccw_req *cqr, *next;
+	struct dasd_device *device;
+	unsigned long long expires;
+	unsigned long flags;
+	u8 int_code, status;
+	addr_t ip;
+	int rc;
+
+	int_code = *((u8 *) DASD_DIAG_LC_INT_CODE);
+	status = *((u8 *) DASD_DIAG_LC_INT_STATUS);
+	switch (int_code) {
+	case DASD_DIAG_CODE_31BIT:
+		ip = (addr_t) *((u32 *) DASD_DIAG_LC_INT_PARM_31BIT);
+		break;
+	case DASD_DIAG_CODE_64BIT:
+		ip = (addr_t) *((u64 *) DASD_DIAG_LC_INT_PARM_64BIT);
+		break;
+	default:
+		return;
+	}
+	if (!ip) {		/* no intparm: unsolicited interrupt */
+		DBF_EVENT(DBF_NOTICE, "%s", "caught unsolicited "
+			      "interrupt");
+		return;
+	}
+	cqr = (struct dasd_ccw_req *) ip;
+	device = (struct dasd_device *) cqr->startdev;
+	if (strncmp(device->discipline->ebcname, (char *) &cqr->magic, 4)) {
+		DBF_DEV_EVENT(DBF_WARNING, device,
+			    " magic number of dasd_ccw_req 0x%08X doesn't"
+			    " match discipline 0x%08X",
+			    cqr->magic, *(int *) (&device->discipline->name));
+		return;
+	}
+
+	/* get irq lock to modify request queue */
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+
+	/* Check for a pending clear operation */
+	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;
+	}
+
+	cqr->stopclk = get_clock();
+
+	expires = 0;
+	if (status == 0) {
+		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, devlist);
+			if (next->status == DASD_CQR_QUEUED) {
+				rc = dasd_start_diag(next);
+				if (rc == 0)
+					expires = next->expires;
+			}
+		}
+	} else {
+		cqr->status = DASD_CQR_QUEUED;
+		DBF_DEV_EVENT(DBF_DEBUG, device, "interrupt status for "
+			    "request %p was %d (%d retries left)", cqr, status,
+			    cqr->retries);
+		dasd_diag_erp(device);
+	}
+
+	if (expires != 0)
+		dasd_device_set_timer(device, expires);
+	else
+		dasd_device_clear_timer(device);
+	dasd_schedule_device_bh(device);
+
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+}
+
+/* Check whether device can be controlled by DIAG discipline. Return zero on
+ * success, non-zero otherwise. */
+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;
+	struct vtoc_cms_label *label;
+	blocknum_t end_block;
+	unsigned int sb, bsize;
+	int rc;
+
+	private = (struct dasd_diag_private *) device->private;
+	if (private == NULL) {
+		private = kzalloc(sizeof(struct dasd_diag_private),GFP_KERNEL);
+		if (private == NULL) {
+			DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+				"Allocating memory for private DASD data "
+				      "failed\n");
+			return -ENOMEM;
+		}
+		ccw_device_get_id(device->cdev, &private->dev_id);
+		device->private = (void *) private;
+	}
+	block = dasd_alloc_block();
+	if (IS_ERR(block)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "could not allocate dasd block structure");
+		device->private = NULL;
+		kfree(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;
+	rdc_data->rdc_len = sizeof (struct dasd_diag_characteristics);
+
+	rc = diag210((struct diag210 *) rdc_data);
+	if (rc) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "failed to retrieve device "
+			    "information (rc=%d)", rc);
+		rc = -EOPNOTSUPP;
+		goto out;
+	}
+
+	/* Figure out position of label block */
+	switch (private->rdc_data.vdev_class) {
+	case DEV_CLASS_FBA:
+		private->pt_block = 1;
+		break;
+	case DEV_CLASS_ECKD:
+		private->pt_block = 2;
+		break;
+	default:
+		dev_warn(&device->cdev->dev, "Device type %d is not supported "
+			    "in DIAG mode\n", private->rdc_data.vdev_class);
+		rc = -EOPNOTSUPP;
+		goto out;
+	}
+
+	DBF_DEV_EVENT(DBF_INFO, device,
+		      "%04X: %04X on real %04X/%02X",
+		      rdc_data->dev_nr,
+		      rdc_data->vdev_type,
+		      rdc_data->rdev_type, rdc_data->rdev_model);
+
+	/* terminate all outstanding operations */
+	mdsk_term_io(device);
+
+	/* figure out blocksize of device */
+	label = (struct vtoc_cms_label *) get_zeroed_page(GFP_KERNEL);
+	if (label == NULL)  {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "No memory to allocate initialization request");
+		rc = -ENOMEM;
+		goto out;
+	}
+	rc = 0;
+	end_block = 0;
+	/* try all sizes - needed for ECKD devices */
+	for (bsize = 512; bsize <= PAGE_SIZE; bsize <<= 1) {
+		mdsk_init_io(device, bsize, 0, &end_block);
+		memset(&bio, 0, sizeof (struct dasd_diag_bio));
+		bio.type = MDSK_READ_REQ;
+		bio.block_number = private->pt_block + 1;
+		bio.buffer = label;
+		memset(&private->iob, 0, sizeof (struct dasd_diag_rw_io));
+		private->iob.dev_nr = rdc_data->dev_nr;
+		private->iob.key = 0;
+		private->iob.flags = 0;	/* do synchronous io */
+		private->iob.block_count = 1;
+		private->iob.interrupt_params = 0;
+		private->iob.bio_list = &bio;
+		private->iob.flaga = DASD_DIAG_FLAGA_DEFAULT;
+		rc = dia250(&private->iob, RW_BIO);
+		if (rc == 3) {
+			dev_warn(&device->cdev->dev,
+				"A 64-bit DIAG call failed\n");
+			rc = -EOPNOTSUPP;
+			goto out_label;
+		}
+		mdsk_term_io(device);
+		if (rc == 0)
+			break;
+	}
+	if (bsize > PAGE_SIZE) {
+		dev_warn(&device->cdev->dev, "Accessing the DASD failed because"
+			 " of an incorrect format (rc=%d)\n", rc);
+		rc = -EIO;
+		goto out_label;
+	}
+	/* check for label block */
+	if (memcmp(label->label_id, DASD_DIAG_CMS1,
+		  sizeof(DASD_DIAG_CMS1)) == 0) {
+		/* get formatted blocksize from label block */
+		bsize = (unsigned int) label->block_size;
+		block->blocks = (unsigned long) label->block_count;
+	} else
+		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)
+		block->s2b_shift++;
+	rc = mdsk_init_io(device, block->bp_block, 0, NULL);
+	if (rc && (rc != 4)) {
+		dev_warn(&device->cdev->dev, "DIAG initialization "
+			"failed with rc=%d\n", rc);
+		rc = -EIO;
+	} else {
+		if (rc == 4)
+			device->features |= DASD_FEATURE_READONLY;
+		dev_info(&device->cdev->dev,
+			 "New DASD with %ld byte/block, total size %ld KB%s\n",
+			 (unsigned long) block->bp_block,
+			 (unsigned long) (block->blocks <<
+					  block->s2b_shift) >> 1,
+			 (rc == 4) ? ", read-only device" : "");
+		rc = 0;
+	}
+out_label:
+	free_page((long) label);
+out:
+	if (rc) {
+		device->block = NULL;
+		dasd_free_block(block);
+		device->private = NULL;
+		kfree(private);
+	}
+	return rc;
+}
+
+/* Fill in virtual disk geometry for device. Return zero on success, non-zero
+ * otherwise. */
+static int
+dasd_diag_fill_geometry(struct dasd_block *block, struct hd_geometry *geo)
+{
+	if (dasd_check_blocksize(block->bp_block) != 0)
+		return -EINVAL;
+	geo->cylinders = (block->blocks << block->s2b_shift) >> 10;
+	geo->heads = 16;
+	geo->sectors = 128 >> block->s2b_shift;
+	return 0;
+}
+
+static dasd_erp_fn_t
+dasd_diag_erp_action(struct dasd_ccw_req * cqr)
+{
+	return dasd_default_erp_action;
+}
+
+static dasd_erp_fn_t
+dasd_diag_erp_postaction(struct dasd_ccw_req * cqr)
+{
+	return dasd_default_erp_postaction;
+}
+
+/* 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 *memdev,
+					       struct dasd_block *block,
+					       struct request *req)
+{
+	struct dasd_ccw_req *cqr;
+	struct dasd_diag_req *dreq;
+	struct dasd_diag_bio *dbio;
+	struct req_iterator iter;
+	struct bio_vec *bv;
+	char *dst;
+	unsigned int count, datasize;
+	sector_t recid, first_rec, last_rec;
+	unsigned int blksize, off;
+	unsigned char rw_cmd;
+
+	if (rq_data_dir(req) == READ)
+		rw_cmd = MDSK_READ_REQ;
+	else if (rq_data_dir(req) == WRITE)
+		rw_cmd = MDSK_WRITE_REQ;
+	else
+		return ERR_PTR(-EINVAL);
+	blksize = block->bp_block;
+	/* Calculate record id of first and last block. */
+	first_rec = blk_rq_pos(req) >> block->s2b_shift;
+	last_rec =
+		(blk_rq_pos(req) + blk_rq_sectors(req) - 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 >> (block->s2b_shift + 9);
+	}
+	/* Paranoia. */
+	if (count != last_rec - first_rec + 1)
+		return ERR_PTR(-EINVAL);
+	/* Build the request */
+	datasize = sizeof(struct dasd_diag_req) +
+		count*sizeof(struct dasd_diag_bio);
+	cqr = dasd_smalloc_request(DASD_DIAG_MAGIC, 0, datasize, memdev);
+	if (IS_ERR(cqr))
+		return cqr;
+
+	dreq = (struct dasd_diag_req *) cqr->data;
+	dreq->block_count = count;
+	dbio = dreq->bio;
+	recid = first_rec;
+	rq_for_each_segment(bv, req, iter) {
+		dst = page_address(bv->bv_page) + bv->bv_offset;
+		for (off = 0; off < bv->bv_len; off += blksize) {
+			memset(dbio, 0, sizeof (struct dasd_diag_bio));
+			dbio->type = rw_cmd;
+			dbio->block_number = recid + 1;
+			dbio->buffer = dst;
+			dbio++;
+			dst += blksize;
+			recid++;
+		}
+	}
+	cqr->retries = DIAG_MAX_RETRIES;
+	cqr->buildclk = get_clock();
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = memdev;
+	cqr->memdev = memdev;
+	cqr->block = block;
+	cqr->expires = DIAG_TIMEOUT;
+	cqr->status = DASD_CQR_FILLED;
+	return cqr;
+}
+
+/* Release DASD request. Return non-zero if request was successful, zero
+ * otherwise. */
+static int
+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->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,
+		    struct dasd_information2_t * info)
+{
+	struct dasd_diag_private *private;
+
+	private = (struct dasd_diag_private *) device->private;
+	info->label_block = (unsigned int) private->pt_block;
+	info->FBA_layout = 1;
+	info->format = DASD_FORMAT_LDL;
+	info->characteristics_size = sizeof (struct dasd_diag_characteristics);
+	memcpy(info->characteristics,
+	       &((struct dasd_diag_private *) device->private)->rdc_data,
+	       sizeof (struct dasd_diag_characteristics));
+	info->confdata_size = 0;
+	return 0;
+}
+
+static void
+dasd_diag_dump_sense(struct dasd_device *device, struct dasd_ccw_req * req,
+		     struct irb *stat)
+{
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+		    "dump sense not available for DIAG data");
+}
+
+static struct dasd_discipline dasd_diag_discipline = {
+	.owner = THIS_MODULE,
+	.name = "DIAG",
+	.ebcname = "DIAG",
+	.max_blocks = DIAG_MAX_BLOCKS,
+	.check_device = dasd_diag_check_device,
+	.fill_geometry = dasd_diag_fill_geometry,
+	.start_IO = dasd_start_diag,
+	.term_IO = dasd_diag_term_IO,
+	.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,
+	.free_cp = dasd_diag_free_cp,
+	.dump_sense = dasd_diag_dump_sense,
+	.fill_info = dasd_diag_fill_info,
+};
+
+static int __init
+dasd_diag_init(void)
+{
+	if (!MACHINE_IS_VM) {
+		pr_info("Discipline %s cannot be used without z/VM\n",
+			dasd_diag_discipline.name);
+		return -ENODEV;
+	}
+	ASCEBC(dasd_diag_discipline.ebcname, 4);
+
+	ctl_set_bit(0, 9);
+	register_external_interrupt(0x2603, dasd_ext_handler);
+	dasd_diag_discipline_pointer = &dasd_diag_discipline;
+	return 0;
+}
+
+static void __exit
+dasd_diag_cleanup(void)
+{
+	unregister_external_interrupt(0x2603, dasd_ext_handler);
+	ctl_clear_bit(0, 9);
+	dasd_diag_discipline_pointer = NULL;
+}
+
+module_init(dasd_diag_init);
+module_exit(dasd_diag_cleanup);

kernel/drivers/s390/block/dasd_diag.h

@@ -0,0 +1,127 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_diag.h
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Based on.......: linux/drivers/s390/block/mdisk.h
+ * ...............: by Hartmunt Penner <hpenner@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000
+ *
+ */
+
+#define MDSK_WRITE_REQ 0x01
+#define MDSK_READ_REQ  0x02
+
+#define INIT_BIO	0x00
+#define RW_BIO		0x01
+#define TERM_BIO	0x02
+
+#define DEV_CLASS_FBA	0x01
+#define DEV_CLASS_ECKD	0x04
+
+#define DASD_DIAG_LC_INT_CODE		132
+#define DASD_DIAG_LC_INT_STATUS		133
+#define DASD_DIAG_LC_INT_PARM_31BIT	128
+#define DASD_DIAG_LC_INT_PARM_64BIT	4536
+#define DASD_DIAG_CODE_31BIT		0x03
+#define DASD_DIAG_CODE_64BIT		0x07
+
+#define DASD_DIAG_RWFLAG_ASYNC		0x02
+#define DASD_DIAG_RWFLAG_NOCACHE	0x01
+
+#define DASD_DIAG_FLAGA_FORMAT_64BIT	0x80
+
+struct dasd_diag_characteristics {
+	u16 dev_nr;
+	u16 rdc_len;
+	u8 vdev_class;
+	u8 vdev_type;
+	u8 vdev_status;
+	u8 vdev_flags;
+	u8 rdev_class;
+	u8 rdev_type;
+	u8 rdev_model;
+	u8 rdev_features;
+} __attribute__ ((packed, aligned(4)));
+
+
+#ifdef CONFIG_64BIT
+#define DASD_DIAG_FLAGA_DEFAULT		DASD_DIAG_FLAGA_FORMAT_64BIT
+
+typedef u64 blocknum_t;
+typedef s64 sblocknum_t;
+
+struct dasd_diag_bio {
+	u8 type;
+	u8 status;
+	u8 spare1[2];
+	u32 alet;
+	blocknum_t block_number;
+	void *buffer;
+} __attribute__ ((packed, aligned(8)));
+
+struct dasd_diag_init_io {
+	u16 dev_nr;
+	u8 flaga;
+	u8 spare1[21];
+	u32 block_size;
+	u8 spare2[4];
+	blocknum_t offset;
+	sblocknum_t start_block;
+	blocknum_t end_block;
+	u8  spare3[8];
+} __attribute__ ((packed, aligned(8)));
+
+struct dasd_diag_rw_io {
+	u16 dev_nr;
+	u8  flaga;
+	u8  spare1[21];
+	u8  key;
+	u8  flags;
+	u8  spare2[2];
+	u32 block_count;
+	u32 alet;
+	u8  spare3[4];
+	u64 interrupt_params;
+	struct dasd_diag_bio *bio_list;
+	u8  spare4[8];
+} __attribute__ ((packed, aligned(8)));
+#else /* CONFIG_64BIT */
+#define DASD_DIAG_FLAGA_DEFAULT		0x0
+
+typedef u32 blocknum_t;
+typedef s32 sblocknum_t;
+
+struct dasd_diag_bio {
+	u8 type;
+	u8 status;
+	u16 spare1;
+	blocknum_t block_number;
+	u32 alet;
+	void *buffer;
+} __attribute__ ((packed, aligned(8)));
+
+struct dasd_diag_init_io {
+	u16 dev_nr;
+	u8 flaga;
+	u8 spare1[21];
+	u32 block_size;
+	blocknum_t offset;
+	sblocknum_t start_block;
+	blocknum_t end_block;
+	u8 spare2[24];
+} __attribute__ ((packed, aligned(8)));
+
+struct dasd_diag_rw_io {
+	u16 dev_nr;
+	u8 flaga;
+	u8 spare1[21];
+	u8 key;
+	u8 flags;
+	u8 spare2[2];
+	u32 block_count;
+	u32 alet;
+	struct dasd_diag_bio *bio_list;
+	u32 interrupt_params;
+	u8 spare3[20];
+} __attribute__ ((packed, aligned(8)));
+#endif /* CONFIG_64BIT */

kernel/drivers/s390/block/dasd_eckd.h

@@ -0,0 +1,464 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_eckd.h
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000
+ *
+ */
+
+#ifndef DASD_ECKD_H
+#define DASD_ECKD_H
+
+/*****************************************************************************
+ * SECTION: CCW Definitions
+ ****************************************************************************/
+#define DASD_ECKD_CCW_WRITE		 0x05
+#define DASD_ECKD_CCW_READ		 0x06
+#define DASD_ECKD_CCW_WRITE_HOME_ADDRESS 0x09
+#define DASD_ECKD_CCW_READ_HOME_ADDRESS	 0x0a
+#define DASD_ECKD_CCW_WRITE_KD		 0x0d
+#define DASD_ECKD_CCW_READ_KD		 0x0e
+#define DASD_ECKD_CCW_ERASE		 0x11
+#define DASD_ECKD_CCW_READ_COUNT	 0x12
+#define DASD_ECKD_CCW_SLCK		 0x14
+#define DASD_ECKD_CCW_WRITE_RECORD_ZERO	 0x15
+#define DASD_ECKD_CCW_READ_RECORD_ZERO	 0x16
+#define DASD_ECKD_CCW_WRITE_CKD		 0x1d
+#define DASD_ECKD_CCW_READ_CKD		 0x1e
+#define DASD_ECKD_CCW_PSF		 0x27
+#define DASD_ECKD_CCW_RSSD		 0x3e
+#define DASD_ECKD_CCW_LOCATE_RECORD	 0x47
+#define DASD_ECKD_CCW_SNSS		 0x54
+#define DASD_ECKD_CCW_DEFINE_EXTENT	 0x63
+#define DASD_ECKD_CCW_WRITE_MT		 0x85
+#define DASD_ECKD_CCW_READ_MT		 0x86
+#define DASD_ECKD_CCW_WRITE_KD_MT	 0x8d
+#define DASD_ECKD_CCW_READ_KD_MT	 0x8e
+#define DASD_ECKD_CCW_RELEASE		 0x94
+#define DASD_ECKD_CCW_READ_CKD_MT	 0x9e
+#define DASD_ECKD_CCW_WRITE_CKD_MT	 0x9d
+#define DASD_ECKD_CCW_WRITE_TRACK_DATA	 0xA5
+#define DASD_ECKD_CCW_READ_TRACK_DATA	 0xA6
+#define DASD_ECKD_CCW_RESERVE		 0xB4
+#define DASD_ECKD_CCW_PFX		 0xE7
+#define DASD_ECKD_CCW_PFX_READ		 0xEA
+#define DASD_ECKD_CCW_RSCK		 0xF9
+
+/*
+ * Perform Subsystem Function / Sub-Orders
+ */
+#define PSF_ORDER_PRSSD 0x18
+#define PSF_ORDER_SSC	0x1D
+
+/*
+ * Size that is reportet for large volumes in the old 16-bit no_cyl field
+ */
+#define LV_COMPAT_CYL 0xFFFE
+
+/*****************************************************************************
+ * SECTION: Type Definitions
+ ****************************************************************************/
+
+struct eckd_count {
+	__u16 cyl;
+	__u16 head;
+	__u8 record;
+	__u8 kl;
+	__u16 dl;
+} __attribute__ ((packed));
+
+struct ch_t {
+	__u16 cyl;
+	__u16 head;
+} __attribute__ ((packed));
+
+struct chs_t {
+	__u16 cyl;
+	__u16 head;
+	__u32 sector;
+} __attribute__ ((packed));
+
+struct chr_t {
+	__u16 cyl;
+	__u16 head;
+	__u8 record;
+} __attribute__ ((packed));
+
+struct geom_t {
+	__u16 cyl;
+	__u16 head;
+	__u32 sector;
+} __attribute__ ((packed));
+
+struct eckd_home {
+	__u8 skip_control[14];
+	__u16 cell_number;
+	__u8 physical_addr[3];
+	__u8 flag;
+	struct ch_t track_addr;
+	__u8 reserved;
+	__u8 key_length;
+	__u8 reserved2[2];
+} __attribute__ ((packed));
+
+struct DE_eckd_data {
+	struct {
+		unsigned char perm:2;	/* Permissions on this extent */
+		unsigned char reserved:1;
+		unsigned char seek:2;	/* Seek control */
+		unsigned char auth:2;	/* Access authorization */
+		unsigned char pci:1;	/* PCI Fetch mode */
+	} __attribute__ ((packed)) mask;
+	struct {
+		unsigned char mode:2;	/* Architecture mode */
+		unsigned char ckd:1;	/* CKD Conversion */
+		unsigned char operation:3;	/* Operation mode */
+		unsigned char cfw:1;	/* Cache fast write */
+		unsigned char dfw:1;	/* DASD fast write */
+	} __attribute__ ((packed)) attributes;
+	__u16 blk_size;		/* Blocksize */
+	__u16 fast_write_id;
+	__u8 ga_additional;	/* Global Attributes Additional */
+	__u8 ga_extended;	/* Global Attributes Extended	*/
+	struct ch_t beg_ext;
+	struct ch_t end_ext;
+	unsigned long long ep_sys_time; /* Ext Parameter - System Time Stamp */
+	__u8 ep_format;        /* Extended Parameter format byte       */
+	__u8 ep_prio;          /* Extended Parameter priority I/O byte */
+	__u8 ep_reserved1;     /* Extended Parameter Reserved	       */
+	__u8 ep_rec_per_track; /* Number of records on a track	       */
+	__u8 ep_reserved[4];   /* Extended Parameter Reserved	       */
+} __attribute__ ((packed));
+
+struct LO_eckd_data {
+	struct {
+		unsigned char orientation:2;
+		unsigned char operation:6;
+	} __attribute__ ((packed)) operation;
+	struct {
+		unsigned char last_bytes_used:1;
+		unsigned char reserved:6;
+		unsigned char read_count_suffix:1;
+	} __attribute__ ((packed)) auxiliary;
+	__u8 unused;
+	__u8 count;
+	struct ch_t seek_addr;
+	struct chr_t search_arg;
+	__u8 sector;
+	__u16 length;
+} __attribute__ ((packed));
+
+struct LRE_eckd_data {
+	struct {
+		unsigned char orientation:2;
+		unsigned char operation:6;
+	} __attribute__ ((packed)) operation;
+	struct {
+		unsigned char length_valid:1;
+		unsigned char length_scope:1;
+		unsigned char imbedded_ccw_valid:1;
+		unsigned char check_bytes:2;
+		unsigned char imbedded_count_valid:1;
+		unsigned char reserved:1;
+		unsigned char read_count_suffix:1;
+	} __attribute__ ((packed)) auxiliary;
+	__u8 imbedded_ccw;
+	__u8 count;
+	struct ch_t seek_addr;
+	struct chr_t search_arg;
+	__u8 sector;
+	__u16 length;
+	__u8 imbedded_count;
+	__u8 extended_operation;
+	__u16 extended_parameter_length;
+	__u8 extended_parameter[0];
+} __attribute__ ((packed));
+
+/* Prefix data for format 0x00 and 0x01 */
+struct PFX_eckd_data {
+	unsigned char format;
+	struct {
+		unsigned char define_extent:1;
+		unsigned char time_stamp:1;
+		unsigned char verify_base:1;
+		unsigned char hyper_pav:1;
+		unsigned char reserved:4;
+	} __attribute__ ((packed)) validity;
+	__u8 base_address;
+	__u8 aux;
+	__u8 base_lss;
+	__u8 reserved[7];
+	struct DE_eckd_data define_extent;
+	struct LRE_eckd_data locate_record;
+} __attribute__ ((packed));
+
+struct dasd_eckd_characteristics {
+	__u16 cu_type;
+	struct {
+		unsigned char support:2;
+		unsigned char async:1;
+		unsigned char reserved:1;
+		unsigned char cache_info:1;
+		unsigned char model:3;
+	} __attribute__ ((packed)) cu_model;
+	__u16 dev_type;
+	__u8 dev_model;
+	struct {
+		unsigned char mult_burst:1;
+		unsigned char RT_in_LR:1;
+		unsigned char reserved1:1;
+		unsigned char RD_IN_LR:1;
+		unsigned char reserved2:4;
+		unsigned char reserved3:8;
+		unsigned char defect_wr:1;
+		unsigned char XRC_supported:1;
+		unsigned char reserved4:1;
+		unsigned char striping:1;
+		unsigned char reserved5:4;
+		unsigned char cfw:1;
+		unsigned char reserved6:2;
+		unsigned char cache:1;
+		unsigned char dual_copy:1;
+		unsigned char dfw:1;
+		unsigned char reset_alleg:1;
+		unsigned char sense_down:1;
+	} __attribute__ ((packed)) facilities;
+	__u8 dev_class;
+	__u8 unit_type;
+	__u16 no_cyl;
+	__u16 trk_per_cyl;
+	__u8 sec_per_trk;
+	__u8 byte_per_track[3];
+	__u16 home_bytes;
+	__u8 formula;
+	union {
+		struct {
+			__u8 f1;
+			__u16 f2;
+			__u16 f3;
+		} __attribute__ ((packed)) f_0x01;
+		struct {
+			__u8 f1;
+			__u8 f2;
+			__u8 f3;
+			__u8 f4;
+			__u8 f5;
+		} __attribute__ ((packed)) f_0x02;
+	} __attribute__ ((packed)) factors;
+	__u16 first_alt_trk;
+	__u16 no_alt_trk;
+	__u16 first_dia_trk;
+	__u16 no_dia_trk;
+	__u16 first_sup_trk;
+	__u16 no_sup_trk;
+	__u8 MDR_ID;
+	__u8 OBR_ID;
+	__u8 director;
+	__u8 rd_trk_set;
+	__u16 max_rec_zero;
+	__u8 reserved1;
+	__u8 RWANY_in_LR;
+	__u8 factor6;
+	__u8 factor7;
+	__u8 factor8;
+	__u8 reserved2[3];
+	__u8 reserved3[6];
+	__u32 long_no_cyl;
+} __attribute__ ((packed));
+
+/* elements of the configuration data */
+struct dasd_ned {
+	struct {
+		__u8 identifier:2;
+		__u8 token_id:1;
+		__u8 sno_valid:1;
+		__u8 subst_sno:1;
+		__u8 recNED:1;
+		__u8 emuNED:1;
+		__u8 reserved:1;
+	} __attribute__ ((packed)) flags;
+	__u8 descriptor;
+	__u8 dev_class;
+	__u8 reserved;
+	__u8 dev_type[6];
+	__u8 dev_model[3];
+	__u8 HDA_manufacturer[3];
+	__u8 HDA_location[2];
+	__u8 HDA_seqno[12];
+	__u8 ID;
+	__u8 unit_addr;
+} __attribute__ ((packed));
+
+struct dasd_sneq {
+	struct {
+		__u8 identifier:2;
+		__u8 reserved:6;
+	} __attribute__ ((packed)) flags;
+	__u8 res1;
+	__u16 format;
+	__u8 res2[4];		/* byte  4- 7 */
+	__u8 sua_flags;		/* byte  8    */
+	__u8 base_unit_addr;	/* byte  9    */
+	__u8 res3[22];		/* byte 10-31 */
+} __attribute__ ((packed));
+
+struct vd_sneq {
+	struct {
+		__u8 identifier:2;
+		__u8 reserved:6;
+	} __attribute__ ((packed)) flags;
+	__u8 res1;
+	__u16 format;
+	__u8 res2[4];	/* byte  4- 7 */
+	__u8 uit[16];	/* byte  8-23 */
+	__u8 res3[8];	/* byte 24-31 */
+} __attribute__ ((packed));
+
+struct dasd_gneq {
+	struct {
+		__u8 identifier:2;
+		__u8 reserved:6;
+	} __attribute__ ((packed)) flags;
+	__u8 reserved[7];
+	__u16 subsystemID;
+	__u8 reserved2[22];
+} __attribute__ ((packed));
+
+struct dasd_eckd_path {
+	__u8 opm;
+	__u8 ppm;
+	__u8 npm;
+};
+
+struct dasd_rssd_features {
+	char feature[256];
+} __attribute__((packed));
+
+
+/*
+ * Perform Subsystem Function - Prepare for Read Subsystem Data
+ */
+struct dasd_psf_prssd_data {
+	unsigned char order;
+	unsigned char flags;
+	unsigned char reserved[4];
+	unsigned char suborder;
+	unsigned char varies[5];
+} __attribute__ ((packed));
+
+/*
+ * Perform Subsystem Function - Set Subsystem Characteristics
+ */
+struct dasd_psf_ssc_data {
+	unsigned char order;
+	unsigned char flags;
+	unsigned char cu_type[4];
+	unsigned char suborder;
+	unsigned char reserved[59];
+} __attribute__((packed));
+
+
+/*
+ * some structures and definitions for alias handling
+ */
+struct dasd_unit_address_configuration {
+	struct {
+		char ua_type;
+		char base_ua;
+	} unit[256];
+} __attribute__((packed));
+
+
+#define MAX_DEVICES_PER_LCU 256
+
+/* flags on the LCU  */
+#define NEED_UAC_UPDATE  0x01
+#define UPDATE_PENDING	0x02
+
+enum pavtype {NO_PAV, BASE_PAV, HYPER_PAV};
+
+
+struct alias_root {
+	struct list_head serverlist;
+	spinlock_t lock;
+};
+
+struct alias_server {
+	struct list_head server;
+	struct dasd_uid uid;
+	struct list_head lculist;
+};
+
+struct summary_unit_check_work_data {
+	char reason;
+	struct dasd_device *device;
+	struct work_struct worker;
+};
+
+struct read_uac_work_data {
+	struct dasd_device *device;
+	struct delayed_work dwork;
+};
+
+struct alias_lcu {
+	struct list_head lcu;
+	struct dasd_uid uid;
+	enum pavtype pav;
+	char flags;
+	spinlock_t lock;
+	struct list_head grouplist;
+	struct list_head active_devices;
+	struct list_head inactive_devices;
+	struct dasd_unit_address_configuration *uac;
+	struct summary_unit_check_work_data suc_data;
+	struct read_uac_work_data ruac_data;
+	struct dasd_ccw_req *rsu_cqr;
+};
+
+struct alias_pav_group {
+	struct list_head group;
+	struct dasd_uid uid;
+	struct alias_lcu *lcu;
+	struct list_head baselist;
+	struct list_head aliaslist;
+	struct dasd_device *next;
+};
+
+
+struct dasd_eckd_private {
+	struct dasd_eckd_characteristics rdc_data;
+	u8 *conf_data;
+	int conf_len;
+	/* pointers to specific parts in the conf_data */
+	struct dasd_ned *ned;
+	struct dasd_sneq *sneq;
+	struct vd_sneq *vdsneq;
+	struct dasd_gneq *gneq;
+
+	struct dasd_eckd_path path_data;
+	struct eckd_count count_area[5];
+	int init_cqr_status;
+	int uses_cdl;
+	struct attrib_data_t attrib;	/* e.g. cache operations */
+	struct dasd_rssd_features features;
+	u32 real_cyl;
+
+	/* alias managemnet */
+	struct dasd_uid uid;
+	struct alias_pav_group *pavgroup;
+	struct alias_lcu *lcu;
+	int count;
+};
+
+
+
+int dasd_alias_make_device_known_to_lcu(struct dasd_device *);
+void dasd_alias_disconnect_device_from_lcu(struct dasd_device *);
+int dasd_alias_add_device(struct dasd_device *);
+int dasd_alias_remove_device(struct dasd_device *);
+struct dasd_device *dasd_alias_get_start_dev(struct dasd_device *);
+void dasd_alias_handle_summary_unit_check(struct dasd_device *, struct irb *);
+void dasd_eckd_reset_ccw_to_base_io(struct dasd_ccw_req *);
+
+#endif				/* DASD_ECKD_H */

kernel/drivers/s390/block/dasd_eer.c

@@ -0,0 +1,712 @@
+/*
+ *  Character device driver for extended error reporting.
+ *
+ *  Copyright (C) 2005 IBM Corporation
+ *  extended error reporting for DASD ECKD devices
+ *  Author(s): Stefan Weinhuber <wein@de.ibm.com>
+ */
+
+#define KMSG_COMPONENT "dasd-eckd"
+
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/device.h>
+#include <linux/poll.h>
+#include <linux/mutex.h>
+#include <linux/smp_lock.h>
+#include <linux/err.h>
+
+#include <asm/uaccess.h>
+#include <asm/atomic.h>
+#include <asm/ebcdic.h>
+
+#include "dasd_int.h"
+#include "dasd_eckd.h"
+
+#ifdef PRINTK_HEADER
+#undef PRINTK_HEADER
+#endif				/* PRINTK_HEADER */
+#define PRINTK_HEADER "dasd(eer):"
+
+/*
+ * SECTION: the internal buffer
+ */
+
+/*
+ * The internal buffer is meant to store obaque blobs of data, so it does
+ * not know of higher level concepts like triggers.
+ * It consists of a number of pages that are used as a ringbuffer. Each data
+ * blob is stored in a simple record that consists of an integer, which
+ * contains the size of the following data, and the data bytes themselfes.
+ *
+ * To allow for multiple independent readers we create one internal buffer
+ * each time the device is opened and destroy the buffer when the file is
+ * closed again. The number of pages used for this buffer is determined by
+ * the module parmeter eer_pages.
+ *
+ * One record can be written to a buffer by using the functions
+ * - dasd_eer_start_record (one time per record to write the size to the
+ *                          buffer and reserve the space for the data)
+ * - dasd_eer_write_buffer (one or more times per record to write the data)
+ * The data can be written in several steps but you will have to compute
+ * the total size up front for the invocation of dasd_eer_start_record.
+ * If the ringbuffer is full, dasd_eer_start_record will remove the required
+ * number of old records.
+ *
+ * A record is typically read in two steps, first read the integer that
+ * specifies the size of the following data, then read the data.
+ * Both can be done by
+ * - dasd_eer_read_buffer
+ *
+ * For all mentioned functions you need to get the bufferlock first and keep
+ * it until a complete record is written or read.
+ *
+ * All information necessary to keep track of an internal buffer is kept in
+ * a struct eerbuffer. The buffer specific to a file pointer is strored in
+ * the private_data field of that file. To be able to write data to all
+ * existing buffers, each buffer is also added to the bufferlist.
+ * If the user does not want to read a complete record in one go, we have to
+ * keep track of the rest of the record. residual stores the number of bytes
+ * that are still to deliver. If the rest of the record is invalidated between
+ * two reads then residual will be set to -1 so that the next read will fail.
+ * All entries in the eerbuffer structure are protected with the bufferlock.
+ * To avoid races between writing to a buffer on the one side and creating
+ * and destroying buffers on the other side, the bufferlock must also be used
+ * to protect the bufferlist.
+ */
+
+static int eer_pages = 5;
+module_param(eer_pages, int, S_IRUGO|S_IWUSR);
+
+struct eerbuffer {
+	struct list_head list;
+	char **buffer;
+	int buffersize;
+	int buffer_page_count;
+	int head;
+        int tail;
+	int residual;
+};
+
+static LIST_HEAD(bufferlist);
+static DEFINE_SPINLOCK(bufferlock);
+static DECLARE_WAIT_QUEUE_HEAD(dasd_eer_read_wait_queue);
+
+/*
+ * How many free bytes are available on the buffer.
+ * Needs to be called with bufferlock held.
+ */
+static int dasd_eer_get_free_bytes(struct eerbuffer *eerb)
+{
+	if (eerb->head < eerb->tail)
+		return eerb->tail - eerb->head - 1;
+	return eerb->buffersize - eerb->head + eerb->tail -1;
+}
+
+/*
+ * How many bytes of buffer space are used.
+ * Needs to be called with bufferlock held.
+ */
+static int dasd_eer_get_filled_bytes(struct eerbuffer *eerb)
+{
+
+	if (eerb->head >= eerb->tail)
+		return eerb->head - eerb->tail;
+	return eerb->buffersize - eerb->tail + eerb->head;
+}
+
+/*
+ * The dasd_eer_write_buffer function just copies count bytes of data
+ * to the buffer. Make sure to call dasd_eer_start_record first, to
+ * make sure that enough free space is available.
+ * Needs to be called with bufferlock held.
+ */
+static void dasd_eer_write_buffer(struct eerbuffer *eerb,
+				  char *data, int count)
+{
+
+	unsigned long headindex,localhead;
+	unsigned long rest, len;
+	char *nextdata;
+
+	nextdata = data;
+	rest = count;
+	while (rest > 0) {
+ 		headindex = eerb->head / PAGE_SIZE;
+ 		localhead = eerb->head % PAGE_SIZE;
+		len = min(rest, PAGE_SIZE - localhead);
+		memcpy(eerb->buffer[headindex]+localhead, nextdata, len);
+		nextdata += len;
+		rest -= len;
+		eerb->head += len;
+		if (eerb->head == eerb->buffersize)
+			eerb->head = 0; /* wrap around */
+		BUG_ON(eerb->head > eerb->buffersize);
+	}
+}
+
+/*
+ * Needs to be called with bufferlock held.
+ */
+static int dasd_eer_read_buffer(struct eerbuffer *eerb, char *data, int count)
+{
+
+	unsigned long tailindex,localtail;
+	unsigned long rest, len, finalcount;
+	char *nextdata;
+
+	finalcount = min(count, dasd_eer_get_filled_bytes(eerb));
+	nextdata = data;
+	rest = finalcount;
+	while (rest > 0) {
+ 		tailindex = eerb->tail / PAGE_SIZE;
+ 		localtail = eerb->tail % PAGE_SIZE;
+		len = min(rest, PAGE_SIZE - localtail);
+		memcpy(nextdata, eerb->buffer[tailindex] + localtail, len);
+		nextdata += len;
+		rest -= len;
+		eerb->tail += len;
+		if (eerb->tail == eerb->buffersize)
+			eerb->tail = 0; /* wrap around */
+		BUG_ON(eerb->tail > eerb->buffersize);
+	}
+	return finalcount;
+}
+
+/*
+ * Whenever you want to write a blob of data to the internal buffer you
+ * have to start by using this function first. It will write the number
+ * of bytes that will be written to the buffer. If necessary it will remove
+ * old records to make room for the new one.
+ * Needs to be called with bufferlock held.
+ */
+static int dasd_eer_start_record(struct eerbuffer *eerb, int count)
+{
+	int tailcount;
+
+	if (count + sizeof(count) > eerb->buffersize)
+		return -ENOMEM;
+	while (dasd_eer_get_free_bytes(eerb) < count + sizeof(count)) {
+		if (eerb->residual > 0) {
+			eerb->tail += eerb->residual;
+			if (eerb->tail >= eerb->buffersize)
+				eerb->tail -= eerb->buffersize;
+			eerb->residual = -1;
+		}
+		dasd_eer_read_buffer(eerb, (char *) &tailcount,
+				     sizeof(tailcount));
+		eerb->tail += tailcount;
+		if (eerb->tail >= eerb->buffersize)
+			eerb->tail -= eerb->buffersize;
+	}
+	dasd_eer_write_buffer(eerb, (char*) &count, sizeof(count));
+
+	return 0;
+};
+
+/*
+ * Release pages that are not used anymore.
+ */
+static void dasd_eer_free_buffer_pages(char **buf, int no_pages)
+{
+	int i;
+
+	for (i = 0; i < no_pages; i++)
+		free_page((unsigned long) buf[i]);
+}
+
+/*
+ * Allocate a new set of memory pages.
+ */
+static int dasd_eer_allocate_buffer_pages(char **buf, int no_pages)
+{
+	int i;
+
+	for (i = 0; i < no_pages; i++) {
+		buf[i] = (char *) get_zeroed_page(GFP_KERNEL);
+		if (!buf[i]) {
+			dasd_eer_free_buffer_pages(buf, i);
+			return -ENOMEM;
+		}
+	}
+	return 0;
+}
+
+/*
+ * SECTION: The extended error reporting functionality
+ */
+
+/*
+ * When a DASD device driver wants to report an error, it calls the
+ * function dasd_eer_write and gives the respective trigger ID as
+ * parameter. Currently there are four kinds of triggers:
+ *
+ * DASD_EER_FATALERROR:  all kinds of unrecoverable I/O problems
+ * DASD_EER_PPRCSUSPEND: PPRC was suspended
+ * DASD_EER_NOPATH:      There is no path to the device left.
+ * DASD_EER_STATECHANGE: The state of the device has changed.
+ *
+ * For the first three triggers all required information can be supplied by
+ * the caller. For these triggers a record is written by the function
+ * dasd_eer_write_standard_trigger.
+ *
+ * The DASD_EER_STATECHANGE trigger is special since a sense subsystem
+ * status ccw need to be executed to gather the necessary sense data first.
+ * The dasd_eer_snss function will queue the SNSS request and the request
+ * callback will then call dasd_eer_write with the DASD_EER_STATCHANGE
+ * trigger.
+ *
+ * To avoid memory allocations at runtime, the necessary memory is allocated
+ * when the extended error reporting is enabled for a device (by
+ * dasd_eer_probe). There is one sense subsystem status request for each
+ * eer enabled DASD device. The presence of the cqr in device->eer_cqr
+ * indicates that eer is enable for the device. The use of the snss request
+ * is protected by the DASD_FLAG_EER_IN_USE bit. When this flag indicates
+ * that the cqr is currently in use, dasd_eer_snss cannot start a second
+ * request but sets the DASD_FLAG_EER_SNSS flag instead. The callback of
+ * the SNSS request will check the bit and call dasd_eer_snss again.
+ */
+
+#define SNSS_DATA_SIZE 44
+
+#define DASD_EER_BUSID_SIZE 10
+struct dasd_eer_header {
+	__u32 total_size;
+	__u32 trigger;
+	__u64 tv_sec;
+	__u64 tv_usec;
+	char busid[DASD_EER_BUSID_SIZE];
+} __attribute__ ((packed));
+
+/*
+ * The following function can be used for those triggers that have
+ * all necessary data available when the function is called.
+ * If the parameter cqr is not NULL, the chain of requests will be searched
+ * for valid sense data, and all valid sense data sets will be added to
+ * the triggers data.
+ */
+static void dasd_eer_write_standard_trigger(struct dasd_device *device,
+					    struct dasd_ccw_req *cqr,
+					    int trigger)
+{
+	struct dasd_ccw_req *temp_cqr;
+	int data_size;
+	struct timeval tv;
+	struct dasd_eer_header header;
+	unsigned long flags;
+	struct eerbuffer *eerb;
+	char *sense;
+
+	/* go through cqr chain and count the valid sense data sets */
+	data_size = 0;
+	for (temp_cqr = cqr; temp_cqr; temp_cqr = temp_cqr->refers)
+		if (dasd_get_sense(&temp_cqr->irb))
+			data_size += 32;
+
+	header.total_size = sizeof(header) + data_size + 4; /* "EOR" */
+	header.trigger = trigger;
+	do_gettimeofday(&tv);
+	header.tv_sec = tv.tv_sec;
+	header.tv_usec = tv.tv_usec;
+	strncpy(header.busid, dev_name(&device->cdev->dev),
+		DASD_EER_BUSID_SIZE);
+
+	spin_lock_irqsave(&bufferlock, flags);
+	list_for_each_entry(eerb, &bufferlist, list) {
+		dasd_eer_start_record(eerb, header.total_size);
+		dasd_eer_write_buffer(eerb, (char *) &header, sizeof(header));
+		for (temp_cqr = cqr; temp_cqr; temp_cqr = temp_cqr->refers) {
+			sense = dasd_get_sense(&temp_cqr->irb);
+			if (sense)
+				dasd_eer_write_buffer(eerb, sense, 32);
+		}
+		dasd_eer_write_buffer(eerb, "EOR", 4);
+	}
+	spin_unlock_irqrestore(&bufferlock, flags);
+	wake_up_interruptible(&dasd_eer_read_wait_queue);
+}
+
+/*
+ * This function writes a DASD_EER_STATECHANGE trigger.
+ */
+static void dasd_eer_write_snss_trigger(struct dasd_device *device,
+					struct dasd_ccw_req *cqr,
+					int trigger)
+{
+	int data_size;
+	int snss_rc;
+	struct timeval tv;
+	struct dasd_eer_header header;
+	unsigned long flags;
+	struct eerbuffer *eerb;
+
+	snss_rc = (cqr->status == DASD_CQR_DONE) ? 0 : -EIO;
+	if (snss_rc)
+		data_size = 0;
+	else
+		data_size = SNSS_DATA_SIZE;
+
+	header.total_size = sizeof(header) + data_size + 4; /* "EOR" */
+	header.trigger = DASD_EER_STATECHANGE;
+	do_gettimeofday(&tv);
+	header.tv_sec = tv.tv_sec;
+	header.tv_usec = tv.tv_usec;
+	strncpy(header.busid, dev_name(&device->cdev->dev),
+		DASD_EER_BUSID_SIZE);
+
+	spin_lock_irqsave(&bufferlock, flags);
+	list_for_each_entry(eerb, &bufferlist, list) {
+		dasd_eer_start_record(eerb, header.total_size);
+		dasd_eer_write_buffer(eerb, (char *) &header , sizeof(header));
+		if (!snss_rc)
+			dasd_eer_write_buffer(eerb, cqr->data, SNSS_DATA_SIZE);
+		dasd_eer_write_buffer(eerb, "EOR", 4);
+	}
+	spin_unlock_irqrestore(&bufferlock, flags);
+	wake_up_interruptible(&dasd_eer_read_wait_queue);
+}
+
+/*
+ * This function is called for all triggers. It calls the appropriate
+ * function that writes the actual trigger records.
+ */
+void dasd_eer_write(struct dasd_device *device, struct dasd_ccw_req *cqr,
+		    unsigned int id)
+{
+	if (!device->eer_cqr)
+		return;
+	switch (id) {
+	case DASD_EER_FATALERROR:
+	case DASD_EER_PPRCSUSPEND:
+		dasd_eer_write_standard_trigger(device, cqr, id);
+		break;
+	case DASD_EER_NOPATH:
+		dasd_eer_write_standard_trigger(device, NULL, id);
+		break;
+	case DASD_EER_STATECHANGE:
+		dasd_eer_write_snss_trigger(device, cqr, id);
+		break;
+	default: /* unknown trigger, so we write it without any sense data */
+		dasd_eer_write_standard_trigger(device, NULL, id);
+		break;
+	}
+}
+EXPORT_SYMBOL(dasd_eer_write);
+
+/*
+ * Start a sense subsystem status request.
+ * Needs to be called with the device held.
+ */
+void dasd_eer_snss(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+
+	cqr = device->eer_cqr;
+	if (!cqr)	/* Device not eer enabled. */
+		return;
+	if (test_and_set_bit(DASD_FLAG_EER_IN_USE, &device->flags)) {
+		/* Sense subsystem status request in use. */
+		set_bit(DASD_FLAG_EER_SNSS, &device->flags);
+		return;
+	}
+	/* cdev is already locked, can't use dasd_add_request_head */
+	clear_bit(DASD_FLAG_EER_SNSS, &device->flags);
+	cqr->status = DASD_CQR_QUEUED;
+	list_add(&cqr->devlist, &device->ccw_queue);
+	dasd_schedule_device_bh(device);
+}
+
+/*
+ * Callback function for use with sense subsystem status request.
+ */
+static void dasd_eer_snss_cb(struct dasd_ccw_req *cqr, void *data)
+{
+	struct dasd_device *device = cqr->startdev;
+	unsigned long flags;
+
+	dasd_eer_write(device, cqr, DASD_EER_STATECHANGE);
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	if (device->eer_cqr == cqr) {
+		clear_bit(DASD_FLAG_EER_IN_USE, &device->flags);
+		if (test_bit(DASD_FLAG_EER_SNSS, &device->flags))
+			/* Another SNSS has been requested in the meantime. */
+			dasd_eer_snss(device);
+		cqr = NULL;
+	}
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	if (cqr)
+		/*
+		 * Extended error recovery has been switched off while
+		 * the SNSS request was running. It could even have
+		 * been switched off and on again in which case there
+		 * is a new ccw in device->eer_cqr. Free the "old"
+		 * snss request now.
+		 */
+		dasd_kfree_request(cqr, device);
+}
+
+/*
+ * Enable error reporting on a given device.
+ */
+int dasd_eer_enable(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+	unsigned long flags;
+	struct ccw1 *ccw;
+
+	if (device->eer_cqr)
+		return 0;
+
+	if (!device->discipline || strcmp(device->discipline->name, "ECKD"))
+		return -EPERM;	/* FIXME: -EMEDIUMTYPE ? */
+
+	cqr = dasd_kmalloc_request(DASD_ECKD_MAGIC, 1 /* SNSS */,
+				   SNSS_DATA_SIZE, device);
+	if (IS_ERR(cqr))
+		return -ENOMEM;
+
+	cqr->startdev = device;
+	cqr->retries = 255;
+	cqr->expires = 10 * HZ;
+	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_SNSS;
+	ccw->count = SNSS_DATA_SIZE;
+	ccw->flags = 0;
+	ccw->cda = (__u32)(addr_t) cqr->data;
+
+	cqr->buildclk = get_clock();
+	cqr->status = DASD_CQR_FILLED;
+	cqr->callback = dasd_eer_snss_cb;
+
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	if (!device->eer_cqr) {
+		device->eer_cqr = cqr;
+		cqr = NULL;
+	}
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	if (cqr)
+		dasd_kfree_request(cqr, device);
+	return 0;
+}
+
+/*
+ * Disable error reporting on a given device.
+ */
+void dasd_eer_disable(struct dasd_device *device)
+{
+	struct dasd_ccw_req *cqr;
+	unsigned long flags;
+	int in_use;
+
+	if (!device->eer_cqr)
+		return;
+	spin_lock_irqsave(get_ccwdev_lock(device->cdev), flags);
+	cqr = device->eer_cqr;
+	device->eer_cqr = NULL;
+	clear_bit(DASD_FLAG_EER_SNSS, &device->flags);
+	in_use = test_and_clear_bit(DASD_FLAG_EER_IN_USE, &device->flags);
+	spin_unlock_irqrestore(get_ccwdev_lock(device->cdev), flags);
+	if (cqr && !in_use)
+		dasd_kfree_request(cqr, device);
+}
+
+/*
+ * SECTION: the device operations
+ */
+
+/*
+ * On the one side we need a lock to access our internal buffer, on the
+ * other side a copy_to_user can sleep. So we need to copy the data we have
+ * to transfer in a readbuffer, which is protected by the readbuffer_mutex.
+ */
+static char readbuffer[PAGE_SIZE];
+static DEFINE_MUTEX(readbuffer_mutex);
+
+static int dasd_eer_open(struct inode *inp, struct file *filp)
+{
+	struct eerbuffer *eerb;
+	unsigned long flags;
+
+	eerb = kzalloc(sizeof(struct eerbuffer), GFP_KERNEL);
+	if (!eerb)
+		return -ENOMEM;
+	lock_kernel();
+	eerb->buffer_page_count = eer_pages;
+	if (eerb->buffer_page_count < 1 ||
+	    eerb->buffer_page_count > INT_MAX / PAGE_SIZE) {
+		kfree(eerb);
+		DBF_EVENT(DBF_WARNING, "can't open device since module "
+			"parameter eer_pages is smaller than 1 or"
+			" bigger than %d", (int)(INT_MAX / PAGE_SIZE));
+		unlock_kernel();
+		return -EINVAL;
+	}
+	eerb->buffersize = eerb->buffer_page_count * PAGE_SIZE;
+	eerb->buffer = kmalloc(eerb->buffer_page_count * sizeof(char *),
+			       GFP_KERNEL);
+        if (!eerb->buffer) {
+		kfree(eerb);
+		unlock_kernel();
+                return -ENOMEM;
+	}
+	if (dasd_eer_allocate_buffer_pages(eerb->buffer,
+					   eerb->buffer_page_count)) {
+		kfree(eerb->buffer);
+		kfree(eerb);
+		unlock_kernel();
+		return -ENOMEM;
+	}
+	filp->private_data = eerb;
+	spin_lock_irqsave(&bufferlock, flags);
+	list_add(&eerb->list, &bufferlist);
+	spin_unlock_irqrestore(&bufferlock, flags);
+
+	unlock_kernel();
+	return nonseekable_open(inp,filp);
+}
+
+static int dasd_eer_close(struct inode *inp, struct file *filp)
+{
+	struct eerbuffer *eerb;
+	unsigned long flags;
+
+	eerb = (struct eerbuffer *) filp->private_data;
+	spin_lock_irqsave(&bufferlock, flags);
+	list_del(&eerb->list);
+	spin_unlock_irqrestore(&bufferlock, flags);
+	dasd_eer_free_buffer_pages(eerb->buffer, eerb->buffer_page_count);
+	kfree(eerb->buffer);
+	kfree(eerb);
+
+	return 0;
+}
+
+static ssize_t dasd_eer_read(struct file *filp, char __user *buf,
+			     size_t count, loff_t *ppos)
+{
+	int tc,rc;
+	int tailcount,effective_count;
+        unsigned long flags;
+	struct eerbuffer *eerb;
+
+	eerb = (struct eerbuffer *) filp->private_data;
+	if (mutex_lock_interruptible(&readbuffer_mutex))
+		return -ERESTARTSYS;
+
+	spin_lock_irqsave(&bufferlock, flags);
+
+	if (eerb->residual < 0) { /* the remainder of this record */
+		                  /* has been deleted             */
+		eerb->residual = 0;
+		spin_unlock_irqrestore(&bufferlock, flags);
+		mutex_unlock(&readbuffer_mutex);
+		return -EIO;
+	} else if (eerb->residual > 0) {
+		/* OK we still have a second half of a record to deliver */
+		effective_count = min(eerb->residual, (int) count);
+		eerb->residual -= effective_count;
+	} else {
+		tc = 0;
+		while (!tc) {
+			tc = dasd_eer_read_buffer(eerb, (char *) &tailcount,
+						  sizeof(tailcount));
+			if (!tc) {
+				/* no data available */
+				spin_unlock_irqrestore(&bufferlock, flags);
+				mutex_unlock(&readbuffer_mutex);
+				if (filp->f_flags & O_NONBLOCK)
+					return -EAGAIN;
+				rc = wait_event_interruptible(
+					dasd_eer_read_wait_queue,
+					eerb->head != eerb->tail);
+				if (rc)
+					return rc;
+				if (mutex_lock_interruptible(&readbuffer_mutex))
+					return -ERESTARTSYS;
+				spin_lock_irqsave(&bufferlock, flags);
+			}
+		}
+		WARN_ON(tc != sizeof(tailcount));
+		effective_count = min(tailcount,(int)count);
+		eerb->residual = tailcount - effective_count;
+	}
+
+	tc = dasd_eer_read_buffer(eerb, readbuffer, effective_count);
+	WARN_ON(tc != effective_count);
+
+	spin_unlock_irqrestore(&bufferlock, flags);
+
+	if (copy_to_user(buf, readbuffer, effective_count)) {
+		mutex_unlock(&readbuffer_mutex);
+		return -EFAULT;
+	}
+
+	mutex_unlock(&readbuffer_mutex);
+	return effective_count;
+}
+
+static unsigned int dasd_eer_poll(struct file *filp, poll_table *ptable)
+{
+	unsigned int mask;
+	unsigned long flags;
+	struct eerbuffer *eerb;
+
+	eerb = (struct eerbuffer *) filp->private_data;
+	poll_wait(filp, &dasd_eer_read_wait_queue, ptable);
+	spin_lock_irqsave(&bufferlock, flags);
+	if (eerb->head != eerb->tail)
+		mask = POLLIN | POLLRDNORM ;
+	else
+		mask = 0;
+	spin_unlock_irqrestore(&bufferlock, flags);
+	return mask;
+}
+
+static const struct file_operations dasd_eer_fops = {
+	.open		= &dasd_eer_open,
+	.release	= &dasd_eer_close,
+	.read		= &dasd_eer_read,
+	.poll		= &dasd_eer_poll,
+	.owner		= THIS_MODULE,
+};
+
+static struct miscdevice *dasd_eer_dev = NULL;
+
+int __init dasd_eer_init(void)
+{
+	int rc;
+
+	dasd_eer_dev = kzalloc(sizeof(*dasd_eer_dev), GFP_KERNEL);
+	if (!dasd_eer_dev)
+		return -ENOMEM;
+
+	dasd_eer_dev->minor = MISC_DYNAMIC_MINOR;
+	dasd_eer_dev->name  = "dasd_eer";
+	dasd_eer_dev->fops  = &dasd_eer_fops;
+
+	rc = misc_register(dasd_eer_dev);
+	if (rc) {
+		kfree(dasd_eer_dev);
+		dasd_eer_dev = NULL;
+		DBF_EVENT(DBF_ERR, "%s", "dasd_eer_init could not "
+		       "register misc device");
+		return rc;
+	}
+
+	return 0;
+}
+
+void dasd_eer_exit(void)
+{
+	if (dasd_eer_dev) {
+		WARN_ON(misc_deregister(dasd_eer_dev) != 0);
+		kfree(dasd_eer_dev);
+		dasd_eer_dev = NULL;
+	}
+}

kernel/drivers/s390/block/dasd_erp.c

@@ -0,0 +1,184 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd.c
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999-2001
+ *
+ */
+
+#define KMSG_COMPONENT "dasd"
+
+#include <linux/ctype.h>
+#include <linux/init.h>
+
+#include <asm/debug.h>
+#include <asm/ebcdic.h>
+#include <asm/uaccess.h>
+
+/* This is ugly... */
+#define PRINTK_HEADER "dasd_erp:"
+
+#include "dasd_int.h"
+
+struct dasd_ccw_req *
+dasd_alloc_erp_request(char *magic, int cplength, int datasize,
+		       struct dasd_device * device)
+{
+	unsigned long flags;
+	struct dasd_ccw_req *cqr;
+	char *data;
+	int size;
+
+	/* Sanity checks */
+	BUG_ON( magic == NULL || datasize > PAGE_SIZE ||
+	     (cplength*sizeof(struct ccw1)) > PAGE_SIZE);
+
+	size = (sizeof(struct dasd_ccw_req) + 7L) & -8L;
+	if (cplength > 0)
+		size += cplength * sizeof(struct ccw1);
+	if (datasize > 0)
+		size += datasize;
+	spin_lock_irqsave(&device->mem_lock, flags);
+	cqr = (struct dasd_ccw_req *)
+		dasd_alloc_chunk(&device->erp_chunks, size);
+	spin_unlock_irqrestore(&device->mem_lock, flags);
+	if (cqr == NULL)
+		return ERR_PTR(-ENOMEM);
+	memset(cqr, 0, sizeof(struct dasd_ccw_req));
+	INIT_LIST_HEAD(&cqr->devlist);
+	INIT_LIST_HEAD(&cqr->blocklist);
+	data = (char *) cqr + ((sizeof(struct dasd_ccw_req) + 7L) & -8L);
+	cqr->cpaddr = NULL;
+	if (cplength > 0) {
+		cqr->cpaddr = (struct ccw1 *) data;
+		data += cplength*sizeof(struct ccw1);
+		memset(cqr->cpaddr, 0, cplength*sizeof(struct ccw1));
+	}
+	cqr->data = NULL;
+	if (datasize > 0) {
+		cqr->data = data;
+ 		memset(cqr->data, 0, datasize);
+	}
+	strncpy((char *) &cqr->magic, magic, 4);
+	ASCEBC((char *) &cqr->magic, 4);
+	set_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
+	dasd_get_device(device);
+	return cqr;
+}
+
+void
+dasd_free_erp_request(struct dasd_ccw_req *cqr, struct dasd_device * device)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&device->mem_lock, flags);
+	dasd_free_chunk(&device->erp_chunks, cqr);
+	spin_unlock_irqrestore(&device->mem_lock, flags);
+	atomic_dec(&device->ref_count);
+}
+
+
+/*
+ * dasd_default_erp_action just retries the current cqr
+ */
+struct dasd_ccw_req *
+dasd_default_erp_action(struct dasd_ccw_req *cqr)
+{
+	struct dasd_device *device;
+
+	device = cqr->startdev;
+
+        /* just retry - there is nothing to save ... I got no sense data.... */
+        if (cqr->retries > 0) {
+		DBF_DEV_EVENT(DBF_DEBUG, device,
+                             "default ERP called (%i retries left)",
+                             cqr->retries);
+		cqr->lpm    = LPM_ANYPATH;
+		cqr->status = DASD_CQR_FILLED;
+        } else {
+		pr_err("%s: default ERP has run out of retries and failed\n",
+		       dev_name(&device->cdev->dev));
+		cqr->status = DASD_CQR_FAILED;
+		cqr->stopclk = get_clock();
+        }
+        return cqr;
+}				/* end dasd_default_erp_action */
+
+/*
+ * DESCRIPTION
+ *   Frees all ERPs of the current ERP Chain and set the status
+ *   of the original CQR either to DASD_CQR_DONE if ERP was successful
+ *   or to DASD_CQR_FAILED if ERP was NOT successful.
+ *   NOTE: This function is only called if no discipline postaction
+ *	   is available
+ *
+ * PARAMETER
+ *   erp		current erp_head
+ *
+ * RETURN VALUES
+ *   cqr		pointer to the original CQR
+ */
+struct dasd_ccw_req *dasd_default_erp_postaction(struct dasd_ccw_req *cqr)
+{
+	int success;
+
+	BUG_ON(cqr->refers == NULL || cqr->function == NULL);
+
+	success = cqr->status == DASD_CQR_DONE;
+
+	/* free all ERPs - but NOT the original cqr */
+	while (cqr->refers != NULL) {
+		struct dasd_ccw_req *refers;
+
+		refers = cqr->refers;
+		/* remove the request from the block queue */
+		list_del(&cqr->blocklist);
+		/* free the finished erp request */
+		dasd_free_erp_request(cqr, cqr->memdev);
+		cqr = refers;
+	}
+
+	/* set corresponding status to original cqr */
+	if (success)
+		cqr->status = DASD_CQR_DONE;
+	else {
+		cqr->status = DASD_CQR_FAILED;
+		cqr->stopclk = get_clock();
+	}
+
+	return cqr;
+
+}				/* end default_erp_postaction */
+
+void
+dasd_log_sense(struct dasd_ccw_req *cqr, struct irb *irb)
+{
+	struct dasd_device *device;
+
+	device = cqr->startdev;
+	/* dump sense data */
+	if (device->discipline && device->discipline->dump_sense)
+		device->discipline->dump_sense(device, cqr, irb);
+}
+
+void
+dasd_log_sense_dbf(struct dasd_ccw_req *cqr, struct irb *irb)
+{
+	struct dasd_device *device;
+
+	device = cqr->startdev;
+	/* dump sense data to s390 debugfeature*/
+	if (device->discipline && device->discipline->dump_sense_dbf)
+		device->discipline->dump_sense_dbf(device, irb, "log");
+}
+EXPORT_SYMBOL(dasd_log_sense_dbf);
+
+EXPORT_SYMBOL(dasd_default_erp_action);
+EXPORT_SYMBOL(dasd_default_erp_postaction);
+EXPORT_SYMBOL(dasd_alloc_erp_request);
+EXPORT_SYMBOL(dasd_free_erp_request);
+EXPORT_SYMBOL(dasd_log_sense);
+

kernel/drivers/s390/block/dasd_fba.c

@@ -0,0 +1,626 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_fba.c
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2009
+ */
+
+#define KMSG_COMPONENT "dasd-fba"
+
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <asm/debug.h>
+
+#include <linux/slab.h>
+#include <linux/hdreg.h>	/* HDIO_GETGEO			    */
+#include <linux/bio.h>
+#include <linux/module.h>
+#include <linux/init.h>
+
+#include <asm/idals.h>
+#include <asm/ebcdic.h>
+#include <asm/io.h>
+#include <asm/todclk.h>
+#include <asm/ccwdev.h>
+
+#include "dasd_int.h"
+#include "dasd_fba.h"
+
+#ifdef PRINTK_HEADER
+#undef PRINTK_HEADER
+#endif				/* PRINTK_HEADER */
+#define PRINTK_HEADER "dasd(fba):"
+
+#define DASD_FBA_CCW_WRITE 0x41
+#define DASD_FBA_CCW_READ 0x42
+#define DASD_FBA_CCW_LOCATE 0x43
+#define DASD_FBA_CCW_DEFINE_EXTENT 0x63
+
+MODULE_LICENSE("GPL");
+
+static struct dasd_discipline dasd_fba_discipline;
+
+struct dasd_fba_private {
+	struct dasd_fba_characteristics rdc_data;
+};
+
+static struct ccw_device_id dasd_fba_ids[] = {
+	{ CCW_DEVICE_DEVTYPE (0x6310, 0, 0x9336, 0), .driver_info = 0x1},
+	{ CCW_DEVICE_DEVTYPE (0x3880, 0, 0x3370, 0), .driver_info = 0x2},
+	{ /* end of list */ },
+};
+
+MODULE_DEVICE_TABLE(ccw, dasd_fba_ids);
+
+static struct ccw_driver dasd_fba_driver; /* see below */
+static int
+dasd_fba_probe(struct ccw_device *cdev)
+{
+	return dasd_generic_probe(cdev, &dasd_fba_discipline);
+}
+
+static int
+dasd_fba_set_online(struct ccw_device *cdev)
+{
+	return dasd_generic_set_online(cdev, &dasd_fba_discipline);
+}
+
+static struct ccw_driver dasd_fba_driver = {
+	.name        = "dasd-fba",
+	.owner       = THIS_MODULE,
+	.ids         = dasd_fba_ids,
+	.probe       = dasd_fba_probe,
+	.remove      = dasd_generic_remove,
+	.set_offline = dasd_generic_set_offline,
+	.set_online  = dasd_fba_set_online,
+	.notify      = dasd_generic_notify,
+	.freeze      = dasd_generic_pm_freeze,
+	.thaw	     = dasd_generic_restore_device,
+	.restore     = dasd_generic_restore_device,
+};
+
+static void
+define_extent(struct ccw1 * ccw, struct DE_fba_data *data, int rw,
+	      int blksize, int beg, int nr)
+{
+	ccw->cmd_code = DASD_FBA_CCW_DEFINE_EXTENT;
+	ccw->flags = 0;
+	ccw->count = 16;
+	ccw->cda = (__u32) __pa(data);
+	memset(data, 0, sizeof (struct DE_fba_data));
+	if (rw == WRITE)
+		(data->mask).perm = 0x0;
+	else if (rw == READ)
+		(data->mask).perm = 0x1;
+	else
+		data->mask.perm = 0x2;
+	data->blk_size = blksize;
+	data->ext_loc = beg;
+	data->ext_end = nr - 1;
+}
+
+static void
+locate_record(struct ccw1 * ccw, struct LO_fba_data *data, int rw,
+	      int block_nr, int block_ct)
+{
+	ccw->cmd_code = DASD_FBA_CCW_LOCATE;
+	ccw->flags = 0;
+	ccw->count = 8;
+	ccw->cda = (__u32) __pa(data);
+	memset(data, 0, sizeof (struct LO_fba_data));
+	if (rw == WRITE)
+		data->operation.cmd = 0x5;
+	else if (rw == READ)
+		data->operation.cmd = 0x6;
+	else
+		data->operation.cmd = 0x8;
+	data->blk_nr = block_nr;
+	data->blk_ct = block_ct;
+}
+
+static int
+dasd_fba_check_characteristics(struct dasd_device *device)
+{
+	struct dasd_block *block;
+	struct dasd_fba_private *private;
+	struct ccw_device *cdev = device->cdev;
+	int rc;
+
+	private = (struct dasd_fba_private *) device->private;
+	if (!private) {
+		private = kzalloc(sizeof(*private), GFP_KERNEL | GFP_DMA);
+		if (!private) {
+			dev_warn(&device->cdev->dev,
+				 "Allocating memory for private DASD "
+				 "data failed\n");
+			return -ENOMEM;
+		}
+		device->private = (void *) private;
+	} else {
+		memset(private, 0, sizeof(*private));
+	}
+	block = dasd_alloc_block();
+	if (IS_ERR(block)) {
+		DBF_EVENT_DEVID(DBF_WARNING, cdev, "%s", "could not allocate "
+				"dasd block structure");
+		device->private = NULL;
+		kfree(private);
+		return PTR_ERR(block);
+	}
+	device->block = block;
+	block->base = device;
+
+	/* Read Device Characteristics */
+	rc = dasd_generic_read_dev_chars(device, DASD_FBA_MAGIC,
+					 &private->rdc_data, 32);
+	if (rc) {
+		DBF_EVENT_DEVID(DBF_WARNING, cdev, "Read device "
+				"characteristics returned error %d", rc);
+		device->block = NULL;
+		dasd_free_block(block);
+		device->private = NULL;
+		kfree(private);
+		return rc;
+	}
+
+	dev_info(&device->cdev->dev,
+		 "New FBA DASD %04X/%02X (CU %04X/%02X) with %d MB "
+		 "and %d B/blk\n",
+		 cdev->id.dev_type,
+		 cdev->id.dev_model,
+		 cdev->id.cu_type,
+		 cdev->id.cu_model,
+		 ((private->rdc_data.blk_bdsa *
+		   (private->rdc_data.blk_size >> 9)) >> 11),
+		 private->rdc_data.blk_size);
+	return 0;
+}
+
+static int dasd_fba_do_analysis(struct dasd_block *block)
+{
+	struct dasd_fba_private *private;
+	int sb, rc;
+
+	private = (struct dasd_fba_private *) block->base->private;
+	rc = dasd_check_blocksize(private->rdc_data.blk_size);
+	if (rc) {
+		DBF_DEV_EVENT(DBF_WARNING, block->base, "unknown blocksize %d",
+			    private->rdc_data.blk_size);
+		return rc;
+	}
+	block->blocks = private->rdc_data.blk_bdsa;
+	block->bp_block = private->rdc_data.blk_size;
+	block->s2b_shift = 0;	/* bits to shift 512 to get a block */
+	for (sb = 512; sb < private->rdc_data.blk_size; sb = sb << 1)
+		block->s2b_shift++;
+	return 0;
+}
+
+static int dasd_fba_fill_geometry(struct dasd_block *block,
+				  struct hd_geometry *geo)
+{
+	if (dasd_check_blocksize(block->bp_block) != 0)
+		return -EINVAL;
+	geo->cylinders = (block->blocks << block->s2b_shift) >> 10;
+	geo->heads = 16;
+	geo->sectors = 128 >> block->s2b_shift;
+	return 0;
+}
+
+static dasd_erp_fn_t
+dasd_fba_erp_action(struct dasd_ccw_req * cqr)
+{
+	return dasd_default_erp_action;
+}
+
+static dasd_erp_fn_t
+dasd_fba_erp_postaction(struct dasd_ccw_req * cqr)
+{
+	if (cqr->function == dasd_default_erp_action)
+		return dasd_default_erp_postaction;
+
+	DBF_DEV_EVENT(DBF_WARNING, cqr->startdev, "unknown ERP action %p",
+		    cqr->function);
+	return NULL;
+}
+
+static void dasd_fba_handle_unsolicited_interrupt(struct dasd_device *device,
+						   struct irb *irb)
+{
+	char mask;
+
+	/* first of all check for state change pending interrupt */
+	mask = DEV_STAT_ATTENTION | DEV_STAT_DEV_END | DEV_STAT_UNIT_EXCEP;
+	if ((irb->scsw.cmd.dstat & mask) == mask) {
+		dasd_generic_handle_state_change(device);
+		return;
+	}
+
+	/* check for unsolicited interrupts */
+	DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+		    "unsolicited interrupt received");
+	device->discipline->dump_sense_dbf(device, irb, "unsolicited");
+	dasd_schedule_device_bh(device);
+	return;
+};
+
+static struct dasd_ccw_req *dasd_fba_build_cp(struct dasd_device * memdev,
+					      struct dasd_block *block,
+					      struct request *req)
+{
+	struct dasd_fba_private *private;
+	unsigned long *idaws;
+	struct LO_fba_data *LO_data;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	struct req_iterator iter;
+	struct bio_vec *bv;
+	char *dst;
+	int count, cidaw, cplength, datasize;
+	sector_t recid, first_rec, last_rec;
+	unsigned int blksize, off;
+	unsigned char cmd;
+
+	private = (struct dasd_fba_private *) block->base->private;
+	if (rq_data_dir(req) == READ) {
+		cmd = DASD_FBA_CCW_READ;
+	} else if (rq_data_dir(req) == WRITE) {
+		cmd = DASD_FBA_CCW_WRITE;
+	} else
+		return ERR_PTR(-EINVAL);
+	blksize = block->bp_block;
+	/* Calculate record id of first and last block. */
+	first_rec = blk_rq_pos(req) >> block->s2b_shift;
+	last_rec =
+		(blk_rq_pos(req) + blk_rq_sectors(req) - 1) >> block->s2b_shift;
+	/* Check struct bio and count the number of blocks for the request. */
+	count = 0;
+	cidaw = 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 >> (block->s2b_shift + 9);
+#if defined(CONFIG_64BIT)
+		if (idal_is_needed (page_address(bv->bv_page), bv->bv_len))
+			cidaw += bv->bv_len / blksize;
+#endif
+	}
+	/* Paranoia. */
+	if (count != last_rec - first_rec + 1)
+		return ERR_PTR(-EINVAL);
+	/* 1x define extent + 1x locate record + number of blocks */
+	cplength = 2 + count;
+	/* 1x define extent + 1x locate record */
+	datasize = sizeof(struct DE_fba_data) + sizeof(struct LO_fba_data) +
+		cidaw * sizeof(unsigned long);
+	/*
+	 * Find out number of additional locate record ccws if the device
+	 * can't do data chaining.
+	 */
+	if (private->rdc_data.mode.bits.data_chain == 0) {
+		cplength += count - 1;
+		datasize += (count - 1)*sizeof(struct LO_fba_data);
+	}
+	/* Allocate the ccw request. */
+	cqr = dasd_smalloc_request(DASD_FBA_MAGIC, cplength, datasize, memdev);
+	if (IS_ERR(cqr))
+		return cqr;
+	ccw = cqr->cpaddr;
+	/* First ccw is define extent. */
+	define_extent(ccw++, cqr->data, rq_data_dir(req),
+		      block->bp_block, blk_rq_pos(req), blk_rq_sectors(req));
+	/* Build locate_record + read/write ccws. */
+	idaws = (unsigned long *) (cqr->data + sizeof(struct DE_fba_data));
+	LO_data = (struct LO_fba_data *) (idaws + cidaw);
+	/* Locate record for all blocks for smart devices. */
+	if (private->rdc_data.mode.bits.data_chain != 0) {
+		ccw[-1].flags |= CCW_FLAG_CC;
+		locate_record(ccw++, LO_data++, rq_data_dir(req), 0, count);
+	}
+	recid = first_rec;
+	rq_for_each_segment(bv, req, iter) {
+		dst = page_address(bv->bv_page) + bv->bv_offset;
+		if (dasd_page_cache) {
+			char *copy = kmem_cache_alloc(dasd_page_cache,
+						      GFP_DMA | __GFP_NOWARN);
+			if (copy && rq_data_dir(req) == WRITE)
+				memcpy(copy + bv->bv_offset, dst, bv->bv_len);
+			if (copy)
+				dst = copy + bv->bv_offset;
+		}
+		for (off = 0; off < bv->bv_len; off += blksize) {
+			/* Locate record for stupid devices. */
+			if (private->rdc_data.mode.bits.data_chain == 0) {
+				ccw[-1].flags |= CCW_FLAG_CC;
+				locate_record(ccw, LO_data++,
+					      rq_data_dir(req),
+					      recid - first_rec, 1);
+				ccw->flags = CCW_FLAG_CC;
+				ccw++;
+			} else {
+				if (recid > first_rec)
+					ccw[-1].flags |= CCW_FLAG_DC;
+				else
+					ccw[-1].flags |= CCW_FLAG_CC;
+			}
+			ccw->cmd_code = cmd;
+			ccw->count = block->bp_block;
+			if (idal_is_needed(dst, blksize)) {
+				ccw->cda = (__u32)(addr_t) idaws;
+				ccw->flags = CCW_FLAG_IDA;
+				idaws = idal_create_words(idaws, dst, blksize);
+			} else {
+				ccw->cda = (__u32)(addr_t) dst;
+				ccw->flags = 0;
+			}
+			ccw++;
+			dst += blksize;
+			recid++;
+		}
+	}
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = memdev;
+	cqr->memdev = memdev;
+	cqr->block = block;
+	cqr->expires = 5 * 60 * HZ;	/* 5 minutes */
+	cqr->retries = 32;
+	cqr->buildclk = get_clock();
+	cqr->status = DASD_CQR_FILLED;
+	return cqr;
+}
+
+static int
+dasd_fba_free_cp(struct dasd_ccw_req *cqr, struct request *req)
+{
+	struct dasd_fba_private *private;
+	struct ccw1 *ccw;
+	struct req_iterator iter;
+	struct bio_vec *bv;
+	char *dst, *cda;
+	unsigned int blksize, off;
+	int status;
+
+	if (!dasd_page_cache)
+		goto out;
+	private = (struct dasd_fba_private *) cqr->block->base->private;
+	blksize = cqr->block->bp_block;
+	ccw = cqr->cpaddr;
+	/* Skip over define extent & locate record. */
+	ccw++;
+	if (private->rdc_data.mode.bits.data_chain != 0)
+		ccw++;
+	rq_for_each_segment(bv, req, iter) {
+		dst = page_address(bv->bv_page) + bv->bv_offset;
+		for (off = 0; off < bv->bv_len; off += blksize) {
+			/* Skip locate record. */
+			if (private->rdc_data.mode.bits.data_chain == 0)
+				ccw++;
+			if (dst) {
+				if (ccw->flags & CCW_FLAG_IDA)
+					cda = *((char **)((addr_t) ccw->cda));
+				else
+					cda = (char *)((addr_t) ccw->cda);
+				if (dst != cda) {
+					if (rq_data_dir(req) == READ)
+						memcpy(dst, cda, bv->bv_len);
+					kmem_cache_free(dasd_page_cache,
+					    (void *)((addr_t)cda & PAGE_MASK));
+				}
+				dst = NULL;
+			}
+			ccw++;
+		}
+	}
+out:
+	status = cqr->status == DASD_CQR_DONE;
+	dasd_sfree_request(cqr, cqr->memdev);
+	return status;
+}
+
+static void dasd_fba_handle_terminated_request(struct dasd_ccw_req *cqr)
+{
+	cqr->status = DASD_CQR_FILLED;
+};
+
+static int
+dasd_fba_fill_info(struct dasd_device * device,
+		   struct dasd_information2_t * info)
+{
+	info->label_block = 1;
+	info->FBA_layout = 1;
+	info->format = DASD_FORMAT_LDL;
+	info->characteristics_size = sizeof(struct dasd_fba_characteristics);
+	memcpy(info->characteristics,
+	       &((struct dasd_fba_private *) device->private)->rdc_data,
+	       sizeof (struct dasd_fba_characteristics));
+	info->confdata_size = 0;
+	return 0;
+}
+
+static void
+dasd_fba_dump_sense_dbf(struct dasd_device *device, struct irb *irb,
+			char *reason)
+{
+	u64 *sense;
+
+	sense = (u64 *) dasd_get_sense(irb);
+	if (sense) {
+		DBF_DEV_EVENT(DBF_EMERG, device,
+			      "%s: %s %02x%02x%02x %016llx %016llx %016llx "
+			      "%016llx", reason,
+			      scsw_is_tm(&irb->scsw) ? "t" : "c",
+			      scsw_cc(&irb->scsw), scsw_cstat(&irb->scsw),
+			      scsw_dstat(&irb->scsw), sense[0], sense[1],
+			      sense[2], sense[3]);
+	} else {
+		DBF_DEV_EVENT(DBF_EMERG, device, "%s",
+			      "SORRY - NO VALID SENSE AVAILABLE\n");
+	}
+}
+
+
+static void
+dasd_fba_dump_sense(struct dasd_device *device, struct dasd_ccw_req * req,
+		    struct irb *irb)
+{
+	char *page;
+	struct ccw1 *act, *end, *last;
+	int len, sl, sct, count;
+
+	page = (char *) get_zeroed_page(GFP_ATOMIC);
+	if (page == NULL) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "No memory to dump sense data");
+		return;
+	}
+	len = sprintf(page, KERN_ERR PRINTK_HEADER
+		      " I/O status report for device %s:\n",
+		      dev_name(&device->cdev->dev));
+	len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+		       " in req: %p CS: 0x%02X DS: 0x%02X\n", req,
+		       irb->scsw.cmd.cstat, irb->scsw.cmd.dstat);
+	len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+		       " device %s: Failing CCW: %p\n",
+		       dev_name(&device->cdev->dev),
+		       (void *) (addr_t) irb->scsw.cmd.cpa);
+	if (irb->esw.esw0.erw.cons) {
+		for (sl = 0; sl < 4; sl++) {
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+				       " Sense(hex) %2d-%2d:",
+				       (8 * sl), ((8 * sl) + 7));
+
+			for (sct = 0; sct < 8; sct++) {
+				len += sprintf(page + len, " %02x",
+					       irb->ecw[8 * sl + sct]);
+			}
+			len += sprintf(page + len, "\n");
+		}
+	} else {
+	        len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " SORRY - NO VALID SENSE AVAILABLE\n");
+	}
+	printk(KERN_ERR "%s", page);
+
+	/* dump the Channel Program */
+	/* print first CCWs (maximum 8) */
+	act = req->cpaddr;
+        for (last = act; last->flags & (CCW_FLAG_CC | CCW_FLAG_DC); last++);
+	end = min(act + 8, last);
+	len = sprintf(page, KERN_ERR PRINTK_HEADER
+		      " Related CP in req: %p\n", req);
+	while (act <= end) {
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " CCW %p: %08X %08X DAT:",
+			       act, ((int *) act)[0], ((int *) act)[1]);
+		for (count = 0; count < 32 && count < act->count;
+		     count += sizeof(int))
+			len += sprintf(page + len, " %08X",
+				       ((int *) (addr_t) act->cda)
+				       [(count>>2)]);
+		len += sprintf(page + len, "\n");
+		act++;
+	}
+	printk(KERN_ERR "%s", page);
+
+
+	/* print failing CCW area */
+	len = 0;
+	if (act <  ((struct ccw1 *)(addr_t) irb->scsw.cmd.cpa) - 2) {
+		act = ((struct ccw1 *)(addr_t) irb->scsw.cmd.cpa) - 2;
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER "......\n");
+	}
+	end = min((struct ccw1 *)(addr_t) irb->scsw.cmd.cpa + 2, last);
+	while (act <= end) {
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " CCW %p: %08X %08X DAT:",
+			       act, ((int *) act)[0], ((int *) act)[1]);
+		for (count = 0; count < 32 && count < act->count;
+		     count += sizeof(int))
+			len += sprintf(page + len, " %08X",
+				       ((int *) (addr_t) act->cda)
+				       [(count>>2)]);
+		len += sprintf(page + len, "\n");
+		act++;
+	}
+
+	/* print last CCWs */
+	if (act <  last - 2) {
+		act = last - 2;
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER "......\n");
+	}
+	while (act <= last) {
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " CCW %p: %08X %08X DAT:",
+			       act, ((int *) act)[0], ((int *) act)[1]);
+		for (count = 0; count < 32 && count < act->count;
+		     count += sizeof(int))
+			len += sprintf(page + len, " %08X",
+				       ((int *) (addr_t) act->cda)
+				       [(count>>2)]);
+		len += sprintf(page + len, "\n");
+		act++;
+	}
+	if (len > 0)
+		printk(KERN_ERR "%s", page);
+	free_page((unsigned long) page);
+}
+
+/*
+ * max_blocks is dependent on the amount of storage that is available
+ * in the static io buffer for each device. Currently each device has
+ * 8192 bytes (=2 pages). For 64 bit one dasd_mchunkt_t structure has
+ * 24 bytes, the struct dasd_ccw_req has 136 bytes and each block can use
+ * up to 16 bytes (8 for the ccw and 8 for the idal pointer). In
+ * addition we have one define extent ccw + 16 bytes of data and a
+ * locate record ccw for each block (stupid devices!) + 16 bytes of data.
+ * That makes:
+ * (8192 - 24 - 136 - 8 - 16) / 40 = 200.2 blocks at maximum.
+ * We want to fit two into the available memory so that we can immediately
+ * start the next request if one finishes off. That makes 100.1 blocks
+ * for one request. Give a little safety and the result is 96.
+ */
+static struct dasd_discipline dasd_fba_discipline = {
+	.owner = THIS_MODULE,
+	.name = "FBA ",
+	.ebcname = "FBA ",
+	.max_blocks = 96,
+	.check_device = dasd_fba_check_characteristics,
+	.do_analysis = dasd_fba_do_analysis,
+	.fill_geometry = dasd_fba_fill_geometry,
+	.start_IO = dasd_start_IO,
+	.term_IO = dasd_term_IO,
+	.handle_terminated_request = dasd_fba_handle_terminated_request,
+	.erp_action = dasd_fba_erp_action,
+	.erp_postaction = dasd_fba_erp_postaction,
+	.handle_unsolicited_interrupt = dasd_fba_handle_unsolicited_interrupt,
+	.build_cp = dasd_fba_build_cp,
+	.free_cp = dasd_fba_free_cp,
+	.dump_sense = dasd_fba_dump_sense,
+	.dump_sense_dbf = dasd_fba_dump_sense_dbf,
+	.fill_info = dasd_fba_fill_info,
+};
+
+static int __init
+dasd_fba_init(void)
+{
+	int ret;
+
+	ASCEBC(dasd_fba_discipline.ebcname, 4);
+	ret = ccw_driver_register(&dasd_fba_driver);
+	if (!ret)
+		wait_for_device_probe();
+
+	return ret;
+}
+
+static void __exit
+dasd_fba_cleanup(void)
+{
+	ccw_driver_unregister(&dasd_fba_driver);
+}
+
+module_init(dasd_fba_init);
+module_exit(dasd_fba_cleanup);

kernel/drivers/s390/block/dasd_fba.h

@@ -0,0 +1,72 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_fba.h
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999,2000
+ *
+ */
+
+#ifndef DASD_FBA_H
+#define DASD_FBA_H
+
+struct DE_fba_data {
+	struct {
+		unsigned char perm:2;	/* Permissions on this extent */
+		unsigned char zero:2;	/* Must be zero */
+		unsigned char da:1;	/* usually zero */
+		unsigned char diag:1;	/* allow diagnose */
+		unsigned char zero2:2;	/* zero */
+	} __attribute__ ((packed)) mask;
+	__u8 zero;		/* Must be zero */
+	__u16 blk_size;		/* Blocksize */
+	__u32 ext_loc;		/* Extent locator */
+	__u32 ext_beg;		/* logical number of block 0 in extent */
+	__u32 ext_end;		/* logocal number of last block in extent */
+} __attribute__ ((packed));
+
+struct LO_fba_data {
+	struct {
+		unsigned char zero:4;
+		unsigned char cmd:4;
+	} __attribute__ ((packed)) operation;
+	__u8 auxiliary;
+	__u16 blk_ct;
+	__u32 blk_nr;
+} __attribute__ ((packed));
+
+struct dasd_fba_characteristics {
+	union {
+		__u8 c;
+		struct {
+			unsigned char reserved:1;
+			unsigned char overrunnable:1;
+			unsigned char burst_byte:1;
+			unsigned char data_chain:1;
+			unsigned char zeros:4;
+		} __attribute__ ((packed)) bits;
+	} __attribute__ ((packed)) mode;
+	union {
+		__u8 c;
+		struct {
+			unsigned char zero0:1;
+			unsigned char removable:1;
+			unsigned char shared:1;
+			unsigned char zero1:1;
+			unsigned char mam:1;
+			unsigned char zeros:3;
+		} __attribute__ ((packed)) bits;
+	} __attribute__ ((packed)) features;
+	__u8 dev_class;
+	__u8 unit_type;
+	__u16 blk_size;
+	__u32 blk_per_cycl;
+	__u32 blk_per_bound;
+	__u32 blk_bdsa;
+	__u32 reserved0;
+	__u16 reserved1;
+	__u16 blk_ce;
+	__u32 reserved2;
+	__u16 reserved3;
+} __attribute__ ((packed));
+
+#endif				/* DASD_FBA_H */

kernel/drivers/s390/block/dasd_genhd.c

@@ -0,0 +1,178 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_genhd.c
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999-2001
+ *
+ * gendisk related functions for the dasd driver.
+ *
+ */
+
+#define KMSG_COMPONENT "dasd"
+
+#include <linux/interrupt.h>
+#include <linux/fs.h>
+#include <linux/blkpg.h>
+
+#include <asm/uaccess.h>
+
+/* This is ugly... */
+#define PRINTK_HEADER "dasd_gendisk:"
+
+#include "dasd_int.h"
+
+/*
+ * Allocate and register gendisk structure for device.
+ */
+int dasd_gendisk_alloc(struct dasd_block *block)
+{
+	struct gendisk *gdp;
+	struct dasd_device *base;
+	int len;
+
+	/* Make sure the minor for this device exists. */
+	base = block->base;
+	if (base->devindex >= DASD_PER_MAJOR)
+		return -EBUSY;
+
+	gdp = alloc_disk(1 << DASD_PARTN_BITS);
+	if (!gdp)
+		return -ENOMEM;
+
+	/* Initialize gendisk structure. */
+	gdp->major = DASD_MAJOR;
+	gdp->first_minor = base->devindex << DASD_PARTN_BITS;
+	gdp->fops = &dasd_device_operations;
+	gdp->driverfs_dev = &base->cdev->dev;
+
+	/*
+	 * Set device name.
+	 *   dasda - dasdz : 26 devices
+	 *   dasdaa - dasdzz : 676 devices, added up = 702
+	 *   dasdaaa - dasdzzz : 17576 devices, added up = 18278
+	 *   dasdaaaa - dasdzzzz : 456976 devices, added up = 475252
+	 */
+	len = sprintf(gdp->disk_name, "dasd");
+	if (base->devindex > 25) {
+		if (base->devindex > 701) {
+			if (base->devindex > 18277)
+			        len += sprintf(gdp->disk_name + len, "%c",
+					       'a'+(((base->devindex-18278)
+						     /17576)%26));
+			len += sprintf(gdp->disk_name + len, "%c",
+				       'a'+(((base->devindex-702)/676)%26));
+		}
+		len += sprintf(gdp->disk_name + len, "%c",
+			       'a'+(((base->devindex-26)/26)%26));
+	}
+	len += sprintf(gdp->disk_name + len, "%c", 'a'+(base->devindex%26));
+
+	if (block->base->features & DASD_FEATURE_READONLY)
+		set_disk_ro(gdp, 1);
+	gdp->private_data = block;
+	gdp->queue = block->request_queue;
+	block->gdp = gdp;
+	set_capacity(block->gdp, 0);
+	add_disk(block->gdp);
+	return 0;
+}
+
+/*
+ * Unregister and free gendisk structure for device.
+ */
+void dasd_gendisk_free(struct dasd_block *block)
+{
+	if (block->gdp) {
+		del_gendisk(block->gdp);
+		block->gdp->queue = NULL;
+		put_disk(block->gdp);
+		block->gdp = NULL;
+	}
+}
+
+/*
+ * Trigger a partition detection.
+ */
+int dasd_scan_partitions(struct dasd_block *block)
+{
+	struct block_device *bdev;
+
+	bdev = bdget_disk(block->gdp, 0);
+	if (!bdev || blkdev_get(bdev, FMODE_READ) < 0)
+		return -ENODEV;
+	/*
+	 * See fs/partition/check.c:register_disk,rescan_partitions
+	 * Can't call rescan_partitions directly. Use ioctl.
+	 */
+	ioctl_by_bdev(bdev, BLKRRPART, 0);
+	/*
+	 * Since the matching blkdev_put call to the blkdev_get in
+	 * this function is not called before dasd_destroy_partitions
+	 * the offline open_count limit needs to be increased from
+	 * 0 to 1. This is done by setting device->bdev (see
+	 * dasd_generic_set_offline). As long as the partition
+	 * detection is running no offline should be allowed. That
+	 * is why the assignment to device->bdev is done AFTER
+	 * the BLKRRPART ioctl.
+	 */
+	block->bdev = bdev;
+	return 0;
+}
+
+/*
+ * Remove all inodes in the system for a device, delete the
+ * partitions and make device unusable by setting its size to zero.
+ */
+void dasd_destroy_partitions(struct dasd_block *block)
+{
+	/* The two structs have 168/176 byte on 31/64 bit. */
+	struct blkpg_partition bpart;
+	struct blkpg_ioctl_arg barg;
+	struct block_device *bdev;
+
+	/*
+	 * Get the bdev pointer from the device structure and clear
+	 * device->bdev to lower the offline open_count limit again.
+	 */
+	bdev = block->bdev;
+	block->bdev = NULL;
+
+	/*
+	 * See fs/partition/check.c:delete_partition
+	 * Can't call delete_partitions directly. Use ioctl.
+	 * The ioctl also does locking and invalidation.
+	 */
+	memset(&bpart, 0, sizeof(struct blkpg_partition));
+	memset(&barg, 0, sizeof(struct blkpg_ioctl_arg));
+	barg.data = (void __force __user *) &bpart;
+	barg.op = BLKPG_DEL_PARTITION;
+	for (bpart.pno = block->gdp->minors - 1; bpart.pno > 0; bpart.pno--)
+		ioctl_by_bdev(bdev, BLKPG, (unsigned long) &barg);
+
+	invalidate_partition(block->gdp, 0);
+	/* Matching blkdev_put to the blkdev_get in dasd_scan_partitions. */
+	blkdev_put(bdev, FMODE_READ);
+	set_capacity(block->gdp, 0);
+}
+
+int dasd_gendisk_init(void)
+{
+	int rc;
+
+	/* Register to static dasd major 94 */
+	rc = register_blkdev(DASD_MAJOR, "dasd");
+	if (rc != 0) {
+		pr_warning("Registering the device driver with major number "
+			   "%d failed\n", DASD_MAJOR);
+		return rc;
+	}
+	return 0;
+}
+
+void dasd_gendisk_exit(void)
+{
+	unregister_blkdev(DASD_MAJOR, "dasd");
+}

kernel/drivers/s390/block/dasd_int.h

@@ -0,0 +1,690 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_int.h
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * Copyright IBM Corp. 1999, 2009
+ */
+
+#ifndef DASD_INT_H
+#define DASD_INT_H
+
+#ifdef __KERNEL__
+
+/* we keep old device allocation scheme; IOW, minors are still in 0..255 */
+#define DASD_PER_MAJOR (1U << (MINORBITS - DASD_PARTN_BITS))
+#define DASD_PARTN_MASK ((1 << DASD_PARTN_BITS) - 1)
+
+/*
+ * States a dasd device can have:
+ *   new: the dasd_device structure is allocated.
+ *   known: the discipline for the device is identified.
+ *   basic: the device can do basic i/o.
+ *   unfmt: the device could not be analyzed (format is unknown).
+ *   ready: partition detection is done and the device is can do block io.
+ *   online: the device accepts requests from the block device queue.
+ *
+ * Things to do for startup state transitions:
+ *   new -> known: find discipline for the device and create devfs entries.
+ *   known -> basic: request irq line for the device.
+ *   basic -> ready: do the initial analysis, e.g. format detection,
+ *                   do block device setup and detect partitions.
+ *   ready -> online: schedule the device tasklet.
+ * Things to do for shutdown state transitions:
+ *   online -> ready: just set the new device state.
+ *   ready -> basic: flush requests from the block device layer, clear
+ *                   partition information and reset format information.
+ *   basic -> known: terminate all requests and free irq.
+ *   known -> new: remove devfs entries and forget discipline.
+ */
+
+#define DASD_STATE_NEW	  0
+#define DASD_STATE_KNOWN  1
+#define DASD_STATE_BASIC  2
+#define DASD_STATE_UNFMT  3
+#define DASD_STATE_READY  4
+#define DASD_STATE_ONLINE 5
+
+#include <linux/module.h>
+#include <linux/wait.h>
+#include <linux/blkdev.h>
+#include <linux/genhd.h>
+#include <linux/hdreg.h>
+#include <linux/interrupt.h>
+#include <linux/log2.h>
+#include <asm/ccwdev.h>
+#include <linux/workqueue.h>
+#include <asm/debug.h>
+#include <asm/dasd.h>
+#include <asm/idals.h>
+
+/* DASD discipline magic */
+#define DASD_ECKD_MAGIC 0xC5C3D2C4
+#define DASD_DIAG_MAGIC 0xC4C9C1C7
+#define DASD_FBA_MAGIC 0xC6C2C140
+
+/*
+ * SECTION: Type definitions
+ */
+struct dasd_device;
+struct dasd_block;
+
+/* BIT DEFINITIONS FOR SENSE DATA */
+#define DASD_SENSE_BIT_0 0x80
+#define DASD_SENSE_BIT_1 0x40
+#define DASD_SENSE_BIT_2 0x20
+#define DASD_SENSE_BIT_3 0x10
+
+/* BIT DEFINITIONS FOR SIM SENSE */
+#define DASD_SIM_SENSE 0x0F
+#define DASD_SIM_MSG_TO_OP 0x03
+#define DASD_SIM_LOG 0x0C
+
+/*
+ * SECTION: MACROs for klogd and s390 debug feature (dbf)
+ */
+#define DBF_DEV_EVENT(d_level, d_device, d_str, d_data...) \
+do { \
+	debug_sprintf_event(d_device->debug_area, \
+			    d_level, \
+			    d_str "\n", \
+			    d_data); \
+} while(0)
+
+#define DBF_DEV_EXC(d_level, d_device, d_str, d_data...) \
+do { \
+	debug_sprintf_exception(d_device->debug_area, \
+				d_level, \
+				d_str "\n", \
+				d_data); \
+} while(0)
+
+#define DBF_EVENT(d_level, d_str, d_data...)\
+do { \
+	debug_sprintf_event(dasd_debug_area, \
+			    d_level,\
+			    d_str "\n", \
+			    d_data); \
+} while(0)
+
+#define DBF_EVENT_DEVID(d_level, d_cdev, d_str, d_data...)	\
+do { \
+	struct ccw_dev_id __dev_id;			\
+	ccw_device_get_id(d_cdev, &__dev_id);		\
+	debug_sprintf_event(dasd_debug_area,		\
+			    d_level,					\
+			    "0.%x.%04x " d_str "\n",			\
+			    __dev_id.ssid, __dev_id.devno, d_data);	\
+} while (0)
+
+#define DBF_EXC(d_level, d_str, d_data...)\
+do { \
+	debug_sprintf_exception(dasd_debug_area, \
+				d_level,\
+				d_str "\n", \
+				d_data); \
+} while(0)
+
+/* limit size for an errorstring */
+#define ERRORLENGTH 30
+
+/* definition of dbf debug levels */
+#define	DBF_EMERG	0	/* system is unusable			*/
+#define	DBF_ALERT	1	/* action must be taken immediately	*/
+#define	DBF_CRIT	2	/* critical conditions			*/
+#define	DBF_ERR		3	/* error conditions			*/
+#define	DBF_WARNING	4	/* warning conditions			*/
+#define	DBF_NOTICE	5	/* normal but significant condition	*/
+#define	DBF_INFO	6	/* informational			*/
+#define	DBF_DEBUG	6	/* debug-level messages			*/
+
+/* messages to be written via klogd and dbf */
+#define DEV_MESSAGE(d_loglevel,d_device,d_string,d_args...)\
+do { \
+	printk(d_loglevel PRINTK_HEADER " %s: " d_string "\n", \
+	       dev_name(&d_device->cdev->dev), d_args); \
+	DBF_DEV_EVENT(DBF_ALERT, d_device, d_string, d_args); \
+} while(0)
+
+#define MESSAGE(d_loglevel,d_string,d_args...)\
+do { \
+	printk(d_loglevel PRINTK_HEADER " " d_string "\n", d_args); \
+	DBF_EVENT(DBF_ALERT, d_string, d_args); \
+} while(0)
+
+/* messages to be written via klogd only */
+#define DEV_MESSAGE_LOG(d_loglevel,d_device,d_string,d_args...)\
+do { \
+	printk(d_loglevel PRINTK_HEADER " %s: " d_string "\n", \
+	       dev_name(&d_device->cdev->dev), d_args); \
+} while(0)
+
+#define MESSAGE_LOG(d_loglevel,d_string,d_args...)\
+do { \
+	printk(d_loglevel PRINTK_HEADER " " d_string "\n", d_args); \
+} while(0)
+
+struct dasd_ccw_req {
+	unsigned int magic;		/* Eye catcher */
+	struct list_head devlist;	/* for dasd_device request queue */
+	struct list_head blocklist;	/* for dasd_block request queue */
+
+	/* Where to execute what... */
+	struct dasd_block *block;	/* the originating block device */
+	struct dasd_device *memdev;	/* the device used to allocate this */
+	struct dasd_device *startdev;	/* device the request is started on */
+	void *cpaddr;			/* address of ccw or tcw */
+	unsigned char cpmode;		/* 0 = cmd mode, 1 = itcw */
+	char status;			/* status of this request */
+	short retries;			/* A retry counter */
+	unsigned long flags;        	/* flags of this request */
+
+	/* ... and how */
+	unsigned long starttime;	/* jiffies time of request start */
+	int expires;			/* expiration period in jiffies */
+	char lpm;			/* logical path mask */
+	void *data;			/* pointer to data area */
+
+	/* these are important for recovering erroneous requests          */
+	int intrc;			/* internal error, e.g. from start_IO */
+	struct irb irb;			/* device status in case of an error */
+	struct dasd_ccw_req *refers;	/* ERP-chain queueing. */
+	void *function; 		/* originating ERP action */
+
+	/* these are for statistics only */
+	unsigned long long buildclk;	/* TOD-clock of request generation */
+	unsigned long long startclk;	/* TOD-clock of request start */
+	unsigned long long stopclk;	/* TOD-clock of request interrupt */
+	unsigned long long endclk;	/* TOD-clock of request termination */
+
+        /* Callback that is called after reaching final status. */
+	void (*callback)(struct dasd_ccw_req *, void *data);
+	void *callback_data;
+};
+
+/*
+ * dasd_ccw_req -> status can be:
+ */
+#define DASD_CQR_FILLED 	0x00	/* request is ready to be processed */
+#define DASD_CQR_DONE		0x01	/* request is completed successfully */
+#define DASD_CQR_NEED_ERP	0x02	/* request needs recovery action */
+#define DASD_CQR_IN_ERP 	0x03	/* request is in recovery */
+#define DASD_CQR_FAILED 	0x04	/* request is finally failed */
+#define DASD_CQR_TERMINATED	0x05	/* request was stopped by driver */
+
+#define DASD_CQR_QUEUED 	0x80	/* request is queued to be processed */
+#define DASD_CQR_IN_IO		0x81	/* request is currently in IO */
+#define DASD_CQR_ERROR		0x82	/* request is completed with error */
+#define DASD_CQR_CLEAR_PENDING	0x83	/* request is clear pending */
+#define DASD_CQR_CLEARED	0x84	/* request was cleared */
+#define DASD_CQR_SUCCESS	0x85	/* request was successful */
+
+
+/* per dasd_ccw_req flags */
+#define DASD_CQR_FLAGS_USE_ERP   0	/* use ERP for this request */
+#define DASD_CQR_FLAGS_FAILFAST  1	/* FAILFAST */
+
+/* Signature for error recovery functions. */
+typedef struct dasd_ccw_req *(*dasd_erp_fn_t) (struct dasd_ccw_req *);
+
+/*
+ * the struct dasd_discipline is
+ * sth like a table of virtual functions, if you think of dasd_eckd
+ * inheriting dasd...
+ * no, currently we are not planning to reimplement the driver in C++
+ */
+struct dasd_discipline {
+	struct module *owner;
+	char ebcname[8];	/* a name used for tagging and printks */
+	char name[8];		/* a name used for tagging and printks */
+	int max_blocks;		/* maximum number of blocks to be chained */
+
+	struct list_head list;	/* used for list of disciplines */
+
+	/*
+	 * Device recognition functions. check_device is used to verify
+	 * the sense data and the information returned by read device
+	 * characteristics. It returns 0 if the discipline can be used
+	 * for the device in question. uncheck_device is called during
+	 * device shutdown to deregister a device from its discipline.
+	 */
+	int (*check_device) (struct dasd_device *);
+	void (*uncheck_device) (struct dasd_device *);
+
+	/*
+	 * do_analysis is used in the step from device state "basic" to
+	 * state "accept". It returns 0 if the device can be made ready,
+	 * it returns -EMEDIUMTYPE if the device can't be made ready or
+	 * -EAGAIN if do_analysis started a ccw that needs to complete
+	 * before the analysis may be repeated.
+	 */
+	int (*do_analysis) (struct dasd_block *);
+
+	/*
+	 * Last things to do when a device is set online, and first things
+	 * when it is set offline.
+	 */
+	int (*ready_to_online) (struct dasd_device *);
+	int (*online_to_ready) (struct dasd_device *);
+
+	/*
+	 * Device operation functions. build_cp creates a ccw chain for
+	 * a block device request, start_io starts the request and
+	 * term_IO cancels it (e.g. in case of a timeout). format_device
+	 * returns a ccw chain to be used to format the device.
+	 * handle_terminated_request allows to examine a cqr and prepare
+	 * it for retry.
+	 */
+	struct dasd_ccw_req *(*build_cp) (struct dasd_device *,
+					  struct dasd_block *,
+					  struct request *);
+	int (*start_IO) (struct dasd_ccw_req *);
+	int (*term_IO) (struct dasd_ccw_req *);
+	void (*handle_terminated_request) (struct dasd_ccw_req *);
+	struct dasd_ccw_req *(*format_device) (struct dasd_device *,
+					       struct format_data_t *);
+	int (*free_cp) (struct dasd_ccw_req *, struct request *);
+
+	/*
+	 * Error recovery functions. examine_error() returns a value that
+	 * indicates what to do for an error condition. If examine_error()
+	 * returns 'dasd_era_recover' erp_action() is called to create a
+	 * special error recovery ccw. erp_postaction() is called after
+	 * an error recovery ccw has finished its execution. dump_sense
+	 * is called for every error condition to print the sense data
+	 * to the console.
+	 */
+	dasd_erp_fn_t(*erp_action) (struct dasd_ccw_req *);
+	dasd_erp_fn_t(*erp_postaction) (struct dasd_ccw_req *);
+	void (*dump_sense) (struct dasd_device *, struct dasd_ccw_req *,
+			    struct irb *);
+	void (*dump_sense_dbf) (struct dasd_device *, struct irb *, char *);
+
+	void (*handle_unsolicited_interrupt) (struct dasd_device *,
+					      struct irb *);
+
+        /* i/o control functions. */
+	int (*fill_geometry) (struct dasd_block *, struct hd_geometry *);
+	int (*fill_info) (struct dasd_device *, struct dasd_information2_t *);
+	int (*ioctl) (struct dasd_block *, unsigned int, void __user *);
+
+	/* suspend/resume functions */
+	int (*freeze) (struct dasd_device *);
+	int (*restore) (struct dasd_device *);
+};
+
+extern struct dasd_discipline *dasd_diag_discipline_pointer;
+
+/*
+ * Unique identifier for dasd device.
+ */
+#define UA_NOT_CONFIGURED  0x00
+#define UA_BASE_DEVICE	   0x01
+#define UA_BASE_PAV_ALIAS  0x02
+#define UA_HYPER_PAV_ALIAS 0x03
+
+struct dasd_uid {
+	__u8 type;
+	char vendor[4];
+	char serial[15];
+	__u16 ssid;
+	__u8 real_unit_addr;
+	__u8 base_unit_addr;
+	char vduit[33];
+};
+
+/*
+ * Notification numbers for extended error reporting notifications:
+ * The DASD_EER_DISABLE notification is sent before a dasd_device (and it's
+ * eer pointer) is freed. The error reporting module needs to do all necessary
+ * cleanup steps.
+ * The DASD_EER_TRIGGER notification sends the actual error reports (triggers).
+ */
+#define DASD_EER_DISABLE 0
+#define DASD_EER_TRIGGER 1
+
+/* Trigger IDs for extended error reporting DASD_EER_TRIGGER notification */
+#define DASD_EER_FATALERROR  1
+#define DASD_EER_NOPATH      2
+#define DASD_EER_STATECHANGE 3
+#define DASD_EER_PPRCSUSPEND 4
+
+struct dasd_device {
+	/* Block device stuff. */
+	struct dasd_block *block;
+
+        unsigned int devindex;
+	unsigned long flags;	   /* per device flags */
+	unsigned short features;   /* copy of devmap-features (read-only!) */
+
+	/* extended error reporting stuff (eer) */
+	struct dasd_ccw_req *eer_cqr;
+
+	/* Device discipline stuff. */
+	struct dasd_discipline *discipline;
+	struct dasd_discipline *base_discipline;
+	char *private;
+
+	/* Device state and target state. */
+	int state, target;
+	int stopped;		/* device (ccw_device_start) was stopped */
+
+	/* reference count. */
+        atomic_t ref_count;
+
+	/* ccw queue and memory for static ccw/erp buffers. */
+	struct list_head ccw_queue;
+	spinlock_t mem_lock;
+	void *ccw_mem;
+	void *erp_mem;
+	struct list_head ccw_chunks;
+	struct list_head erp_chunks;
+
+	atomic_t tasklet_scheduled;
+        struct tasklet_struct tasklet;
+	struct work_struct kick_work;
+	struct work_struct restore_device;
+	struct timer_list timer;
+
+	debug_info_t *debug_area;
+
+	struct ccw_device *cdev;
+
+	/* hook for alias management */
+	struct list_head alias_list;
+};
+
+struct dasd_block {
+	/* Block device stuff. */
+	struct gendisk *gdp;
+	struct request_queue *request_queue;
+	spinlock_t request_queue_lock;
+	struct block_device *bdev;
+	atomic_t open_count;
+
+	unsigned long long blocks; /* size of volume in blocks */
+	unsigned int bp_block;	   /* bytes per block */
+	unsigned int s2b_shift;	   /* log2 (bp_block/512) */
+
+	struct dasd_device *base;
+	struct list_head ccw_queue;
+	spinlock_t queue_lock;
+
+	atomic_t tasklet_scheduled;
+	struct tasklet_struct tasklet;
+	struct timer_list timer;
+
+#ifdef CONFIG_DASD_PROFILE
+	struct dasd_profile_info_t profile;
+#endif
+};
+
+
+
+/* reasons why device (ccw_device_start) was stopped */
+#define DASD_STOPPED_NOT_ACC 1         /* not accessible */
+#define DASD_STOPPED_QUIESCE 2         /* Quiesced */
+#define DASD_STOPPED_PENDING 4         /* long busy */
+#define DASD_STOPPED_DC_WAIT 8         /* disconnected, wait */
+#define DASD_STOPPED_SU      16        /* summary unit check handling */
+#define DASD_STOPPED_PM      32        /* pm state transition */
+#define DASD_UNRESUMED_PM    64        /* pm resume failed state */
+
+/* per device flags */
+#define DASD_FLAG_OFFLINE	3	/* device is in offline processing */
+#define DASD_FLAG_EER_SNSS	4	/* A SNSS is required */
+#define DASD_FLAG_EER_IN_USE	5	/* A SNSS request is running */
+
+void dasd_put_device_wake(struct dasd_device *);
+
+/*
+ * Reference count inliners
+ */
+static inline void
+dasd_get_device(struct dasd_device *device)
+{
+	atomic_inc(&device->ref_count);
+}
+
+static inline void
+dasd_put_device(struct dasd_device *device)
+{
+	if (atomic_dec_return(&device->ref_count) == 0)
+		dasd_put_device_wake(device);
+}
+
+/*
+ * The static memory in ccw_mem and erp_mem is managed by a sorted
+ * list of free memory chunks.
+ */
+struct dasd_mchunk
+{
+	struct list_head list;
+	unsigned long size;
+} __attribute__ ((aligned(8)));
+
+static inline void
+dasd_init_chunklist(struct list_head *chunk_list, void *mem,
+		    unsigned long size)
+{
+	struct dasd_mchunk *chunk;
+
+	INIT_LIST_HEAD(chunk_list);
+	chunk = (struct dasd_mchunk *) mem;
+	chunk->size = size - sizeof(struct dasd_mchunk);
+	list_add(&chunk->list, chunk_list);
+}
+
+static inline void *
+dasd_alloc_chunk(struct list_head *chunk_list, unsigned long size)
+{
+	struct dasd_mchunk *chunk, *tmp;
+
+	size = (size + 7L) & -8L;
+	list_for_each_entry(chunk, chunk_list, list) {
+		if (chunk->size < size)
+			continue;
+		if (chunk->size > size + sizeof(struct dasd_mchunk)) {
+			char *endaddr = (char *) (chunk + 1) + chunk->size;
+			tmp = (struct dasd_mchunk *) (endaddr - size) - 1;
+			tmp->size = size;
+			chunk->size -= size + sizeof(struct dasd_mchunk);
+			chunk = tmp;
+		} else
+			list_del(&chunk->list);
+		return (void *) (chunk + 1);
+	}
+	return NULL;
+}
+
+static inline void
+dasd_free_chunk(struct list_head *chunk_list, void *mem)
+{
+	struct dasd_mchunk *chunk, *tmp;
+	struct list_head *p, *left;
+
+	chunk = (struct dasd_mchunk *)
+		((char *) mem - sizeof(struct dasd_mchunk));
+	/* Find out the left neighbour in chunk_list. */
+	left = chunk_list;
+	list_for_each(p, chunk_list) {
+		if (list_entry(p, struct dasd_mchunk, list) > chunk)
+			break;
+		left = p;
+	}
+	/* Try to merge with right neighbour = next element from left. */
+	if (left->next != chunk_list) {
+		tmp = list_entry(left->next, struct dasd_mchunk, list);
+		if ((char *) (chunk + 1) + chunk->size == (char *) tmp) {
+			list_del(&tmp->list);
+			chunk->size += tmp->size + sizeof(struct dasd_mchunk);
+		}
+	}
+	/* Try to merge with left neighbour. */
+	if (left != chunk_list) {
+		tmp = list_entry(left, struct dasd_mchunk, list);
+		if ((char *) (tmp + 1) + tmp->size == (char *) chunk) {
+			tmp->size += chunk->size + sizeof(struct dasd_mchunk);
+			return;
+		}
+	}
+	__list_add(&chunk->list, left, left->next);
+}
+
+/*
+ * Check if bsize is in { 512, 1024, 2048, 4096 }
+ */
+static inline int
+dasd_check_blocksize(int bsize)
+{
+	if (bsize < 512 || bsize > 4096 || !is_power_of_2(bsize))
+		return -EMEDIUMTYPE;
+	return 0;
+}
+
+/* externals in dasd.c */
+#define DASD_PROFILE_ON	 1
+#define DASD_PROFILE_OFF 0
+
+extern debug_info_t *dasd_debug_area;
+extern struct dasd_profile_info_t dasd_global_profile;
+extern unsigned int dasd_profile_level;
+extern const struct block_device_operations dasd_device_operations;
+
+extern struct kmem_cache *dasd_page_cache;
+
+struct dasd_ccw_req *
+dasd_kmalloc_request(int , int, int, struct dasd_device *);
+struct dasd_ccw_req *
+dasd_smalloc_request(int , int, int, struct dasd_device *);
+void dasd_kfree_request(struct dasd_ccw_req *, struct dasd_device *);
+void dasd_sfree_request(struct dasd_ccw_req *, struct dasd_device *);
+
+static inline int
+dasd_kmalloc_set_cda(struct ccw1 *ccw, void *cda, struct dasd_device *device)
+{
+	return set_normalized_cda(ccw, cda);
+}
+
+struct dasd_device *dasd_alloc_device(void);
+void dasd_free_device(struct dasd_device *);
+
+struct dasd_block *dasd_alloc_block(void);
+void dasd_free_block(struct dasd_block *);
+
+void dasd_enable_device(struct dasd_device *);
+void dasd_set_target_state(struct dasd_device *, int);
+void dasd_kick_device(struct dasd_device *);
+void dasd_restore_device(struct dasd_device *);
+
+void dasd_add_request_head(struct dasd_ccw_req *);
+void dasd_add_request_tail(struct dasd_ccw_req *);
+int  dasd_start_IO(struct dasd_ccw_req *);
+int  dasd_term_IO(struct dasd_ccw_req *);
+void dasd_schedule_device_bh(struct dasd_device *);
+void dasd_schedule_block_bh(struct dasd_block *);
+int  dasd_sleep_on(struct dasd_ccw_req *);
+int  dasd_sleep_on_immediatly(struct dasd_ccw_req *);
+int  dasd_sleep_on_interruptible(struct dasd_ccw_req *);
+void dasd_device_set_timer(struct dasd_device *, int);
+void dasd_device_clear_timer(struct dasd_device *);
+void dasd_block_set_timer(struct dasd_block *, int);
+void dasd_block_clear_timer(struct dasd_block *);
+int  dasd_cancel_req(struct dasd_ccw_req *);
+int dasd_flush_device_queue(struct dasd_device *);
+int dasd_generic_probe (struct ccw_device *, struct dasd_discipline *);
+void dasd_generic_remove (struct ccw_device *cdev);
+int dasd_generic_set_online(struct ccw_device *, struct dasd_discipline *);
+int dasd_generic_set_offline (struct ccw_device *cdev);
+int dasd_generic_notify(struct ccw_device *, int);
+void dasd_generic_handle_state_change(struct dasd_device *);
+int dasd_generic_pm_freeze(struct ccw_device *);
+int dasd_generic_restore_device(struct ccw_device *);
+
+int dasd_generic_read_dev_chars(struct dasd_device *, int, void *, int);
+char *dasd_get_sense(struct irb *);
+
+/* externals in dasd_devmap.c */
+extern int dasd_max_devindex;
+extern int dasd_probeonly;
+extern int dasd_autodetect;
+extern int dasd_nopav;
+extern int dasd_nofcx;
+
+int dasd_devmap_init(void);
+void dasd_devmap_exit(void);
+
+struct dasd_device *dasd_create_device(struct ccw_device *);
+void dasd_delete_device(struct dasd_device *);
+
+int dasd_get_uid(struct ccw_device *, struct dasd_uid *);
+int dasd_set_uid(struct ccw_device *, struct dasd_uid *);
+int dasd_get_feature(struct ccw_device *, int);
+int dasd_set_feature(struct ccw_device *, int, int);
+
+int dasd_add_sysfs_files(struct ccw_device *);
+void dasd_remove_sysfs_files(struct ccw_device *);
+
+struct dasd_device *dasd_device_from_cdev(struct ccw_device *);
+struct dasd_device *dasd_device_from_cdev_locked(struct ccw_device *);
+struct dasd_device *dasd_device_from_devindex(int);
+
+int dasd_parse(void);
+int dasd_busid_known(const char *);
+
+/* externals in dasd_gendisk.c */
+int  dasd_gendisk_init(void);
+void dasd_gendisk_exit(void);
+int dasd_gendisk_alloc(struct dasd_block *);
+void dasd_gendisk_free(struct dasd_block *);
+int dasd_scan_partitions(struct dasd_block *);
+void dasd_destroy_partitions(struct dasd_block *);
+
+/* externals in dasd_ioctl.c */
+int  dasd_ioctl(struct block_device *, fmode_t, unsigned int, unsigned long);
+
+/* externals in dasd_proc.c */
+int dasd_proc_init(void);
+void dasd_proc_exit(void);
+
+/* externals in dasd_erp.c */
+struct dasd_ccw_req *dasd_default_erp_action(struct dasd_ccw_req *);
+struct dasd_ccw_req *dasd_default_erp_postaction(struct dasd_ccw_req *);
+struct dasd_ccw_req *dasd_alloc_erp_request(char *, int, int,
+					    struct dasd_device *);
+void dasd_free_erp_request(struct dasd_ccw_req *, struct dasd_device *);
+void dasd_log_sense(struct dasd_ccw_req *, struct irb *);
+void dasd_log_sense_dbf(struct dasd_ccw_req *cqr, struct irb *irb);
+
+/* externals in dasd_3990_erp.c */
+struct dasd_ccw_req *dasd_3990_erp_action(struct dasd_ccw_req *);
+void dasd_3990_erp_handle_sim(struct dasd_device *, char *);
+
+/* externals in dasd_eer.c */
+#ifdef CONFIG_DASD_EER
+int dasd_eer_init(void);
+void dasd_eer_exit(void);
+int dasd_eer_enable(struct dasd_device *);
+void dasd_eer_disable(struct dasd_device *);
+void dasd_eer_write(struct dasd_device *, struct dasd_ccw_req *cqr,
+		    unsigned int id);
+void dasd_eer_snss(struct dasd_device *);
+
+static inline int dasd_eer_enabled(struct dasd_device *device)
+{
+	return device->eer_cqr != NULL;
+}
+#else
+#define dasd_eer_init()		(0)
+#define dasd_eer_exit()		do { } while (0)
+#define dasd_eer_enable(d)	(0)
+#define dasd_eer_disable(d)	do { } while (0)
+#define dasd_eer_write(d,c,i)	do { } while (0)
+#define dasd_eer_snss(d)	do { } while (0)
+#define dasd_eer_enabled(d)	(0)
+#endif	/* CONFIG_DASD_ERR */
+
+#endif				/* __KERNEL__ */
+
+#endif				/* DASD_H */

kernel/drivers/s390/block/dasd_ioctl.c

@@ -0,0 +1,436 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_ioctl.c
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999-2001
+ *
+ * i/o controls for the dasd driver.
+ */
+
+#define KMSG_COMPONENT "dasd"
+
+#include <linux/interrupt.h>
+#include <linux/major.h>
+#include <linux/fs.h>
+#include <linux/blkpg.h>
+#include <linux/smp_lock.h>
+
+#include <asm/ccwdev.h>
+#include <asm/cmb.h>
+#include <asm/uaccess.h>
+
+/* This is ugly... */
+#define PRINTK_HEADER "dasd_ioctl:"
+
+#include "dasd_int.h"
+
+
+static int
+dasd_ioctl_api_version(void __user *argp)
+{
+	int ver = DASD_API_VERSION;
+	return put_user(ver, (int __user *)argp);
+}
+
+/*
+ * Enable device.
+ * used by dasdfmt after BIODASDDISABLE to retrigger blocksize detection
+ */
+static int
+dasd_ioctl_enable(struct block_device *bdev)
+{
+	struct dasd_block *block = bdev->bd_disk->private_data;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	dasd_enable_device(block->base);
+	/* Formatting the dasd device can change the capacity. */
+	mutex_lock(&bdev->bd_mutex);
+	i_size_write(bdev->bd_inode, (loff_t)get_capacity(block->gdp) << 9);
+	mutex_unlock(&bdev->bd_mutex);
+	return 0;
+}
+
+/*
+ * Disable device.
+ * Used by dasdfmt. Disable I/O operations but allow ioctls.
+ */
+static int
+dasd_ioctl_disable(struct block_device *bdev)
+{
+	struct dasd_block *block = bdev->bd_disk->private_data;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	/*
+	 * Man this is sick. We don't do a real disable but only downgrade
+	 * the device to DASD_STATE_BASIC. The reason is that dasdfmt uses
+	 * BIODASDDISABLE to disable accesses to the device via the block
+	 * device layer but it still wants to do i/o on the device by
+	 * using the BIODASDFMT ioctl. Therefore the correct state for the
+	 * device is DASD_STATE_BASIC that allows to do basic i/o.
+	 */
+	dasd_set_target_state(block->base, DASD_STATE_BASIC);
+	/*
+	 * Set i_size to zero, since read, write, etc. check against this
+	 * value.
+	 */
+	mutex_lock(&bdev->bd_mutex);
+	i_size_write(bdev->bd_inode, 0);
+	mutex_unlock(&bdev->bd_mutex);
+	return 0;
+}
+
+/*
+ * Quiesce device.
+ */
+static int dasd_ioctl_quiesce(struct dasd_block *block)
+{
+	unsigned long flags;
+	struct dasd_device *base;
+
+	base = block->base;
+	if (!capable (CAP_SYS_ADMIN))
+		return -EACCES;
+
+	pr_info("%s: The DASD has been put in the quiesce "
+		"state\n", dev_name(&base->cdev->dev));
+	spin_lock_irqsave(get_ccwdev_lock(base->cdev), flags);
+	base->stopped |= DASD_STOPPED_QUIESCE;
+	spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags);
+	return 0;
+}
+
+
+/*
+ * Resume device.
+ */
+static int dasd_ioctl_resume(struct dasd_block *block)
+{
+	unsigned long flags;
+	struct dasd_device *base;
+
+	base = block->base;
+	if (!capable (CAP_SYS_ADMIN))
+		return -EACCES;
+
+	pr_info("%s: I/O operations have been resumed "
+		"on the DASD\n", dev_name(&base->cdev->dev));
+	spin_lock_irqsave(get_ccwdev_lock(base->cdev), flags);
+	base->stopped &= ~DASD_STOPPED_QUIESCE;
+	spin_unlock_irqrestore(get_ccwdev_lock(base->cdev), flags);
+
+	dasd_schedule_block_bh(block);
+	return 0;
+}
+
+/*
+ * performs formatting of _device_ according to _fdata_
+ * Note: The discipline's format_function is assumed to deliver formatting
+ * commands to format a single unit of the device. In terms of the ECKD
+ * devices this means CCWs are generated to format a single track.
+ */
+static int dasd_format(struct dasd_block *block, struct format_data_t *fdata)
+{
+	struct dasd_ccw_req *cqr;
+	struct dasd_device *base;
+	int rc;
+
+	base = block->base;
+	if (base->discipline->format_device == NULL)
+		return -EPERM;
+
+	if (base->state != DASD_STATE_BASIC) {
+		pr_warning("%s: The DASD cannot be formatted while it is "
+			   "enabled\n",  dev_name(&base->cdev->dev));
+		return -EBUSY;
+	}
+
+	DBF_DEV_EVENT(DBF_NOTICE, base,
+		      "formatting units %u to %u (%u B blocks) flags %u",
+		      fdata->start_unit,
+		      fdata->stop_unit, fdata->blksize, fdata->intensity);
+
+	/* Since dasdfmt keeps the device open after it was disabled,
+	 * there still exists an inode for this device.
+	 * We must update i_blkbits, otherwise we might get errors when
+	 * enabling the device later.
+	 */
+	if (fdata->start_unit == 0) {
+		struct block_device *bdev = bdget_disk(block->gdp, 0);
+		bdev->bd_inode->i_blkbits = blksize_bits(fdata->blksize);
+		bdput(bdev);
+	}
+
+	while (fdata->start_unit <= fdata->stop_unit) {
+		cqr = base->discipline->format_device(base, fdata);
+		if (IS_ERR(cqr))
+			return PTR_ERR(cqr);
+		rc = dasd_sleep_on_interruptible(cqr);
+		dasd_sfree_request(cqr, cqr->memdev);
+		if (rc) {
+			if (rc != -ERESTARTSYS)
+				pr_err("%s: Formatting unit %d failed with "
+				       "rc=%d\n", dev_name(&base->cdev->dev),
+				       fdata->start_unit, rc);
+			return rc;
+		}
+		fdata->start_unit++;
+	}
+	return 0;
+}
+
+/*
+ * Format device.
+ */
+static int
+dasd_ioctl_format(struct block_device *bdev, void __user *argp)
+{
+	struct dasd_block *block = bdev->bd_disk->private_data;
+	struct format_data_t fdata;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (!argp)
+		return -EINVAL;
+
+	if (block->base->features & DASD_FEATURE_READONLY)
+		return -EROFS;
+	if (copy_from_user(&fdata, argp, sizeof(struct format_data_t)))
+		return -EFAULT;
+	if (bdev != bdev->bd_contains) {
+		pr_warning("%s: The specified DASD is a partition and cannot "
+			   "be formatted\n",
+			   dev_name(&block->base->cdev->dev));
+		return -EINVAL;
+	}
+	return dasd_format(block, &fdata);
+}
+
+#ifdef CONFIG_DASD_PROFILE
+/*
+ * Reset device profile information
+ */
+static int dasd_ioctl_reset_profile(struct dasd_block *block)
+{
+	memset(&block->profile, 0, sizeof(struct dasd_profile_info_t));
+	return 0;
+}
+
+/*
+ * Return device profile information
+ */
+static int dasd_ioctl_read_profile(struct dasd_block *block, void __user *argp)
+{
+	if (dasd_profile_level == DASD_PROFILE_OFF)
+		return -EIO;
+	if (copy_to_user(argp, &block->profile,
+			 sizeof(struct dasd_profile_info_t)))
+		return -EFAULT;
+	return 0;
+}
+#else
+static int dasd_ioctl_reset_profile(struct dasd_block *block)
+{
+	return -ENOSYS;
+}
+
+static int dasd_ioctl_read_profile(struct dasd_block *block, void __user *argp)
+{
+	return -ENOSYS;
+}
+#endif
+
+/*
+ * Return dasd information. Used for BIODASDINFO and BIODASDINFO2.
+ */
+static int dasd_ioctl_information(struct dasd_block *block,
+				  unsigned int cmd, void __user *argp)
+{
+	struct dasd_information2_t *dasd_info;
+	unsigned long flags;
+	int rc;
+	struct dasd_device *base;
+	struct ccw_device *cdev;
+	struct ccw_dev_id dev_id;
+
+	base = block->base;
+	if (!base->discipline || !base->discipline->fill_info)
+		return -EINVAL;
+
+	dasd_info = kzalloc(sizeof(struct dasd_information2_t), GFP_KERNEL);
+	if (dasd_info == NULL)
+		return -ENOMEM;
+
+	rc = base->discipline->fill_info(base, dasd_info);
+	if (rc) {
+		kfree(dasd_info);
+		return rc;
+	}
+
+	cdev = base->cdev;
+	ccw_device_get_id(cdev, &dev_id);
+
+	dasd_info->devno = dev_id.devno;
+	dasd_info->schid = _ccw_device_get_subchannel_number(base->cdev);
+	dasd_info->cu_type = cdev->id.cu_type;
+	dasd_info->cu_model = cdev->id.cu_model;
+	dasd_info->dev_type = cdev->id.dev_type;
+	dasd_info->dev_model = cdev->id.dev_model;
+	dasd_info->status = base->state;
+	/*
+	 * The open_count is increased for every opener, that includes
+	 * the blkdev_get in dasd_scan_partitions.
+	 * This must be hidden from user-space.
+	 */
+	dasd_info->open_count = atomic_read(&block->open_count);
+	if (!block->bdev)
+		dasd_info->open_count++;
+
+	/*
+	 * check if device is really formatted
+	 * LDL / CDL was returned by 'fill_info'
+	 */
+	if ((base->state < DASD_STATE_READY) ||
+	    (dasd_check_blocksize(block->bp_block)))
+		dasd_info->format = DASD_FORMAT_NONE;
+
+	dasd_info->features |=
+		((base->features & DASD_FEATURE_READONLY) != 0);
+
+	memcpy(dasd_info->type, base->discipline->name, 4);
+
+	if (block->request_queue->request_fn) {
+		struct list_head *l;
+#ifdef DASD_EXTENDED_PROFILING
+		{
+			struct list_head *l;
+			spin_lock_irqsave(&block->lock, flags);
+			list_for_each(l, &block->request_queue->queue_head)
+				dasd_info->req_queue_len++;
+			spin_unlock_irqrestore(&block->lock, flags);
+		}
+#endif				/* DASD_EXTENDED_PROFILING */
+		spin_lock_irqsave(get_ccwdev_lock(base->cdev), flags);
+		list_for_each(l, &base->ccw_queue)
+			dasd_info->chanq_len++;
+		spin_unlock_irqrestore(get_ccwdev_lock(base->cdev),
+				       flags);
+	}
+
+	rc = 0;
+	if (copy_to_user(argp, dasd_info,
+			 ((cmd == (unsigned int) BIODASDINFO2) ?
+			  sizeof(struct dasd_information2_t) :
+			  sizeof(struct dasd_information_t))))
+		rc = -EFAULT;
+	kfree(dasd_info);
+	return rc;
+}
+
+/*
+ * Set read only
+ */
+static int
+dasd_ioctl_set_ro(struct block_device *bdev, void __user *argp)
+{
+	struct dasd_block *block =  bdev->bd_disk->private_data;
+	int intval;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+	if (bdev != bdev->bd_contains)
+		// ro setting is not allowed for partitions
+		return -EINVAL;
+	if (get_user(intval, (int __user *)argp))
+		return -EFAULT;
+
+	set_disk_ro(bdev->bd_disk, intval);
+	return dasd_set_feature(block->base->cdev, DASD_FEATURE_READONLY, intval);
+}
+
+static int dasd_ioctl_readall_cmb(struct dasd_block *block, unsigned int cmd,
+		unsigned long arg)
+{
+	struct cmbdata __user *argp = (void __user *) arg;
+	size_t size = _IOC_SIZE(cmd);
+	struct cmbdata data;
+	int ret;
+
+	ret = cmf_readall(block->base->cdev, &data);
+	if (!ret && copy_to_user(argp, &data, min(size, sizeof(*argp))))
+		return -EFAULT;
+	return ret;
+}
+
+static int
+dasd_do_ioctl(struct block_device *bdev, fmode_t mode,
+	      unsigned int cmd, unsigned long arg)
+{
+	struct dasd_block *block = bdev->bd_disk->private_data;
+	void __user *argp = (void __user *)arg;
+
+	if (!block)
+                return -ENODEV;
+
+	if ((_IOC_DIR(cmd) != _IOC_NONE) && !arg) {
+		PRINT_DEBUG("empty data ptr");
+		return -EINVAL;
+	}
+
+	switch (cmd) {
+	case BIODASDDISABLE:
+		return dasd_ioctl_disable(bdev);
+	case BIODASDENABLE:
+		return dasd_ioctl_enable(bdev);
+	case BIODASDQUIESCE:
+		return dasd_ioctl_quiesce(block);
+	case BIODASDRESUME:
+		return dasd_ioctl_resume(block);
+	case BIODASDFMT:
+		return dasd_ioctl_format(bdev, argp);
+	case BIODASDINFO:
+		return dasd_ioctl_information(block, cmd, argp);
+	case BIODASDINFO2:
+		return dasd_ioctl_information(block, cmd, argp);
+	case BIODASDPRRD:
+		return dasd_ioctl_read_profile(block, argp);
+	case BIODASDPRRST:
+		return dasd_ioctl_reset_profile(block);
+	case BLKROSET:
+		return dasd_ioctl_set_ro(bdev, argp);
+	case DASDAPIVER:
+		return dasd_ioctl_api_version(argp);
+	case BIODASDCMFENABLE:
+		return enable_cmf(block->base->cdev);
+	case BIODASDCMFDISABLE:
+		return disable_cmf(block->base->cdev);
+	case BIODASDREADALLCMB:
+		return dasd_ioctl_readall_cmb(block, cmd, arg);
+	default:
+		/* if the discipline has an ioctl method try it. */
+		if (block->base->discipline->ioctl) {
+			int rval = block->base->discipline->ioctl(block, cmd, argp);
+			if (rval != -ENOIOCTLCMD)
+				return rval;
+		}
+
+		return -EINVAL;
+	}
+}
+
+int dasd_ioctl(struct block_device *bdev, fmode_t mode,
+	       unsigned int cmd, unsigned long arg)
+{
+	int rc;
+
+	lock_kernel();
+	rc = dasd_do_ioctl(bdev, mode, cmd, arg);
+	unlock_kernel();
+	return rc;
+}

kernel/drivers/s390/block/dasd_proc.c

@@ -0,0 +1,349 @@
+/*
+ * File...........: linux/drivers/s390/block/dasd_proc.c
+ * Author(s)......: Holger Smolinski <Holger.Smolinski@de.ibm.com>
+ *		    Horst Hummel <Horst.Hummel@de.ibm.com>
+ *		    Carsten Otte <Cotte@de.ibm.com>
+ *		    Martin Schwidefsky <schwidefsky@de.ibm.com>
+ * Bugreports.to..: <Linux390@de.ibm.com>
+ * (C) IBM Corporation, IBM Deutschland Entwicklung GmbH, 1999-2002
+ *
+ * /proc interface for the dasd driver.
+ *
+ */
+
+#define KMSG_COMPONENT "dasd"
+
+#include <linux/ctype.h>
+#include <linux/seq_file.h>
+#include <linux/vmalloc.h>
+#include <linux/proc_fs.h>
+
+#include <asm/debug.h>
+#include <asm/uaccess.h>
+
+/* This is ugly... */
+#define PRINTK_HEADER "dasd_proc:"
+
+#include "dasd_int.h"
+
+static struct proc_dir_entry *dasd_proc_root_entry = NULL;
+static struct proc_dir_entry *dasd_devices_entry = NULL;
+static struct proc_dir_entry *dasd_statistics_entry = NULL;
+
+#ifdef CONFIG_DASD_PROFILE
+static char *
+dasd_get_user_string(const char __user *user_buf, size_t user_len)
+{
+	char *buffer;
+
+	buffer = kmalloc(user_len + 1, GFP_KERNEL);
+	if (buffer == NULL)
+		return ERR_PTR(-ENOMEM);
+	if (copy_from_user(buffer, user_buf, user_len) != 0) {
+		kfree(buffer);
+		return ERR_PTR(-EFAULT);
+	}
+	/* got the string, now strip linefeed. */
+	if (buffer[user_len - 1] == '\n')
+		buffer[user_len - 1] = 0;
+	else
+		buffer[user_len] = 0;
+	return buffer;
+}
+#endif /* CONFIG_DASD_PROFILE */
+
+static int
+dasd_devices_show(struct seq_file *m, void *v)
+{
+	struct dasd_device *device;
+	struct dasd_block *block;
+	char *substr;
+
+	device = dasd_device_from_devindex((unsigned long) v - 1);
+	if (IS_ERR(device))
+		return 0;
+	if (device->block)
+		block = device->block;
+	else {
+		dasd_put_device(device);
+		return 0;
+	}
+	/* Print device number. */
+	seq_printf(m, "%s", dev_name(&device->cdev->dev));
+	/* Print discipline string. */
+	if (device->discipline != NULL)
+		seq_printf(m, "(%s)", device->discipline->name);
+	else
+		seq_printf(m, "(none)");
+	/* Print kdev. */
+	if (block->gdp)
+		seq_printf(m, " at (%3d:%6d)",
+			   MAJOR(disk_devt(block->gdp)),
+			   MINOR(disk_devt(block->gdp)));
+	else
+		seq_printf(m, "  at (???:??????)");
+	/* Print device name. */
+	if (block->gdp)
+		seq_printf(m, " is %-8s", block->gdp->disk_name);
+	else
+		seq_printf(m, " is ????????");
+	/* Print devices features. */
+	substr = (device->features & DASD_FEATURE_READONLY) ? "(ro)" : " ";
+	seq_printf(m, "%4s: ", substr);
+	/* Print device status information. */
+	switch (device->state) {
+	case DASD_STATE_NEW:
+		seq_printf(m, "new");
+		break;
+	case DASD_STATE_KNOWN:
+		seq_printf(m, "detected");
+		break;
+	case DASD_STATE_BASIC:
+		seq_printf(m, "basic");
+		break;
+	case DASD_STATE_UNFMT:
+		seq_printf(m, "unformatted");
+		break;
+	case DASD_STATE_READY:
+	case DASD_STATE_ONLINE:
+		seq_printf(m, "active ");
+		if (dasd_check_blocksize(block->bp_block))
+			seq_printf(m, "n/f	 ");
+		else
+			seq_printf(m,
+				   "at blocksize: %d, %lld blocks, %lld MB",
+				   block->bp_block, block->blocks,
+				   ((block->bp_block >> 9) *
+				    block->blocks) >> 11);
+		break;
+	default:
+		seq_printf(m, "no stat");
+		break;
+	}
+	dasd_put_device(device);
+	if (dasd_probeonly)
+		seq_printf(m, "(probeonly)");
+	seq_printf(m, "\n");
+	return 0;
+}
+
+static void *dasd_devices_start(struct seq_file *m, loff_t *pos)
+{
+	if (*pos >= dasd_max_devindex)
+		return NULL;
+	return (void *)((unsigned long) *pos + 1);
+}
+
+static void *dasd_devices_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return dasd_devices_start(m, pos);
+}
+
+static void dasd_devices_stop(struct seq_file *m, void *v)
+{
+}
+
+static const struct seq_operations dasd_devices_seq_ops = {
+	.start		= dasd_devices_start,
+	.next		= dasd_devices_next,
+	.stop		= dasd_devices_stop,
+	.show		= dasd_devices_show,
+};
+
+static int dasd_devices_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &dasd_devices_seq_ops);
+}
+
+static const struct file_operations dasd_devices_file_ops = {
+	.owner		= THIS_MODULE,
+	.open		= dasd_devices_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static int
+dasd_calc_metrics(char *page, char **start, off_t off,
+		  int count, int *eof, int len)
+{
+	len = (len > off) ? len - off : 0;
+	if (len > count)
+		len = count;
+	if (len < count)
+		*eof = 1;
+	*start = page + off;
+	return len;
+}
+
+#ifdef CONFIG_DASD_PROFILE
+static char *
+dasd_statistics_array(char *str, unsigned int *array, int factor)
+{
+	int i;
+
+	for (i = 0; i < 32; i++) {
+		str += sprintf(str, "%7d ", array[i] / factor);
+		if (i == 15)
+			str += sprintf(str, "\n");
+	}
+	str += sprintf(str,"\n");
+	return str;
+}
+#endif /* CONFIG_DASD_PROFILE */
+
+static int
+dasd_statistics_read(char *page, char **start, off_t off,
+		     int count, int *eof, void *data)
+{
+	unsigned long len;
+#ifdef CONFIG_DASD_PROFILE
+	struct dasd_profile_info_t *prof;
+	char *str;
+	int factor;
+
+	/* check for active profiling */
+	if (dasd_profile_level == DASD_PROFILE_OFF) {
+		len = sprintf(page, "Statistics are off - they might be "
+				    "switched on using 'echo set on > "
+				    "/proc/dasd/statistics'\n");
+		return dasd_calc_metrics(page, start, off, count, eof, len);
+	}
+
+	prof = &dasd_global_profile;
+	/* prevent couter 'overflow' on output */
+	for (factor = 1; (prof->dasd_io_reqs / factor) > 9999999;
+	     factor *= 10);
+
+	str = page;
+	str += sprintf(str, "%d dasd I/O requests\n", prof->dasd_io_reqs);
+	str += sprintf(str, "with %u sectors(512B each)\n",
+		       prof->dasd_io_sects);
+	str += sprintf(str, "Scale Factor is  %d\n", factor);
+	str += sprintf(str,
+		       "   __<4	   ___8	   __16	   __32	   __64	   _128	"
+		       "   _256	   _512	   __1k	   __2k	   __4k	   __8k	"
+		       "   _16k	   _32k	   _64k	   128k\n");
+	str += sprintf(str,
+		       "   _256	   _512	   __1M	   __2M	   __4M	   __8M	"
+		       "   _16M	   _32M	   _64M	   128M	   256M	   512M	"
+		       "   __1G	   __2G	   __4G " "   _>4G\n");
+
+	str += sprintf(str, "Histogram of sizes (512B secs)\n");
+	str = dasd_statistics_array(str, prof->dasd_io_secs, factor);
+	str += sprintf(str, "Histogram of I/O times (microseconds)\n");
+	str = dasd_statistics_array(str, prof->dasd_io_times, factor);
+	str += sprintf(str, "Histogram of I/O times per sector\n");
+	str = dasd_statistics_array(str, prof->dasd_io_timps, factor);
+	str += sprintf(str, "Histogram of I/O time till ssch\n");
+	str = dasd_statistics_array(str, prof->dasd_io_time1, factor);
+	str += sprintf(str, "Histogram of I/O time between ssch and irq\n");
+	str = dasd_statistics_array(str, prof->dasd_io_time2, factor);
+	str += sprintf(str, "Histogram of I/O time between ssch "
+			    "and irq per sector\n");
+	str = dasd_statistics_array(str, prof->dasd_io_time2ps, factor);
+	str += sprintf(str, "Histogram of I/O time between irq and end\n");
+	str = dasd_statistics_array(str, prof->dasd_io_time3, factor);
+	str += sprintf(str, "# of req in chanq at enqueuing (1..32) \n");
+	str = dasd_statistics_array(str, prof->dasd_io_nr_req, factor);
+	len = str - page;
+#else
+	len = sprintf(page, "Statistics are not activated in this kernel\n");
+#endif
+	return dasd_calc_metrics(page, start, off, count, eof, len);
+}
+
+static int
+dasd_statistics_write(struct file *file, const char __user *user_buf,
+		      unsigned long user_len, void *data)
+{
+#ifdef CONFIG_DASD_PROFILE
+	char *buffer, *str;
+
+	if (user_len > 65536)
+		user_len = 65536;
+	buffer = dasd_get_user_string(user_buf, user_len);
+	if (IS_ERR(buffer))
+		return PTR_ERR(buffer);
+	DBF_EVENT(DBF_DEBUG, "/proc/dasd/statictics: '%s'\n", buffer);
+
+	/* check for valid verbs */
+	for (str = buffer; isspace(*str); str++);
+	if (strncmp(str, "set", 3) == 0 && isspace(str[3])) {
+		/* 'set xxx' was given */
+		for (str = str + 4; isspace(*str); str++);
+		if (strcmp(str, "on") == 0) {
+			/* switch on statistics profiling */
+			dasd_profile_level = DASD_PROFILE_ON;
+			pr_info("The statistics feature has been switched "
+				"on\n");
+		} else if (strcmp(str, "off") == 0) {
+			/* switch off and reset statistics profiling */
+			memset(&dasd_global_profile,
+			       0, sizeof (struct dasd_profile_info_t));
+			dasd_profile_level = DASD_PROFILE_OFF;
+			pr_info("The statistics feature has been switched "
+				"off\n");
+		} else
+			goto out_error;
+	} else if (strncmp(str, "reset", 5) == 0) {
+		/* reset the statistics */
+		memset(&dasd_global_profile, 0,
+		       sizeof (struct dasd_profile_info_t));
+		pr_info("The statistics have been reset\n");
+	} else
+		goto out_error;
+	kfree(buffer);
+	return user_len;
+out_error:
+	pr_warning("%s is not a supported value for /proc/dasd/statistics\n",
+		str);
+	kfree(buffer);
+	return -EINVAL;
+#else
+	pr_warning("/proc/dasd/statistics: is not activated in this kernel\n");
+	return user_len;
+#endif				/* CONFIG_DASD_PROFILE */
+}
+
+/*
+ * Create dasd proc-fs entries.
+ * In case creation failed, cleanup and return -ENOENT.
+ */
+int
+dasd_proc_init(void)
+{
+	dasd_proc_root_entry = proc_mkdir("dasd", NULL);
+	if (!dasd_proc_root_entry)
+		goto out_nodasd;
+	dasd_devices_entry = proc_create("devices",
+					 S_IFREG | S_IRUGO | S_IWUSR,
+					 dasd_proc_root_entry,
+					 &dasd_devices_file_ops);
+	if (!dasd_devices_entry)
+		goto out_nodevices;
+	dasd_statistics_entry = create_proc_entry("statistics",
+						  S_IFREG | S_IRUGO | S_IWUSR,
+						  dasd_proc_root_entry);
+	if (!dasd_statistics_entry)
+		goto out_nostatistics;
+	dasd_statistics_entry->read_proc = dasd_statistics_read;
+	dasd_statistics_entry->write_proc = dasd_statistics_write;
+	return 0;
+
+ out_nostatistics:
+	remove_proc_entry("devices", dasd_proc_root_entry);
+ out_nodevices:
+	remove_proc_entry("dasd", NULL);
+ out_nodasd:
+	return -ENOENT;
+}
+
+void
+dasd_proc_exit(void)
+{
+	remove_proc_entry("devices", dasd_proc_root_entry);
+	remove_proc_entry("statistics", dasd_proc_root_entry);
+	remove_proc_entry("dasd", NULL);
+}

kernel/drivers/s390/block/xpram.c

@@ -0,0 +1,476 @@
+/*
+ * Xpram.c -- the S/390 expanded memory RAM-disk
+ *           
+ * significant parts of this code are based on
+ * the sbull device driver presented in
+ * A. Rubini: Linux Device Drivers
+ *
+ * Author of XPRAM specific coding: Reinhard Buendgen
+ *                                  buendgen@de.ibm.com
+ * Rewrite for 2.5: Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *
+ * External interfaces:
+ *   Interfaces to linux kernel
+ *        xpram_setup: read kernel parameters
+ *   Device specific file operations
+ *        xpram_iotcl
+ *        xpram_open
+ *
+ * "ad-hoc" partitioning:
+ *    the expanded memory can be partitioned among several devices 
+ *    (with different minors). The partitioning set up can be
+ *    set by kernel or module parameters (int devs & int sizes[])
+ *
+ * Potential future improvements:
+ *   generic hard disk support to replace ad-hoc partitioning
+ */
+
+#define KMSG_COMPONENT "xpram"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/ctype.h>  /* isdigit, isxdigit */
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/blkdev.h>
+#include <linux/blkpg.h>
+#include <linux/hdreg.h>  /* HDIO_GETGEO */
+#include <linux/sysdev.h>
+#include <linux/bio.h>
+#include <linux/suspend.h>
+#include <linux/platform_device.h>
+#include <asm/uaccess.h>
+
+#define XPRAM_NAME	"xpram"
+#define XPRAM_DEVS	1	/* one partition */
+#define XPRAM_MAX_DEVS	32	/* maximal number of devices (partitions) */
+
+typedef struct {
+	unsigned int	size;		/* size of xpram segment in pages */
+	unsigned int	offset;		/* start page of xpram segment */
+} xpram_device_t;
+
+static xpram_device_t xpram_devices[XPRAM_MAX_DEVS];
+static unsigned int xpram_sizes[XPRAM_MAX_DEVS];
+static struct gendisk *xpram_disks[XPRAM_MAX_DEVS];
+static struct request_queue *xpram_queues[XPRAM_MAX_DEVS];
+static unsigned int xpram_pages;
+static int xpram_devs;
+
+/*
+ * Parameter parsing functions.
+ */
+static int __initdata devs = XPRAM_DEVS;
+static char __initdata *sizes[XPRAM_MAX_DEVS];
+
+module_param(devs, int, 0);
+module_param_array(sizes, charp, NULL, 0);
+
+MODULE_PARM_DESC(devs, "number of devices (\"partitions\"), " \
+		 "the default is " __MODULE_STRING(XPRAM_DEVS) "\n");
+MODULE_PARM_DESC(sizes, "list of device (partition) sizes " \
+		 "the defaults are 0s \n" \
+		 "All devices with size 0 equally partition the "
+		 "remaining space on the expanded strorage not "
+		 "claimed by explicit sizes\n");
+MODULE_LICENSE("GPL");
+
+/*
+ * Copy expanded memory page (4kB) into main memory                  
+ * Arguments                                                         
+ *           page_addr:    address of target page                    
+ *           xpage_index:  index of expandeded memory page           
+ * Return value                                                      
+ *           0:            if operation succeeds
+ *           -EIO:         if pgin failed
+ *           -ENXIO:       if xpram has vanished
+ */
+static int xpram_page_in (unsigned long page_addr, unsigned int xpage_index)
+{
+	int cc = 2;	/* return unused cc 2 if pgin traps */
+
+	asm volatile(
+		"	.insn	rre,0xb22e0000,%1,%2\n"  /* pgin %1,%2 */
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
+		"1:\n"
+		EX_TABLE(0b,1b)
+		: "+d" (cc) : "a" (__pa(page_addr)), "d" (xpage_index) : "cc");
+	if (cc == 3)
+		return -ENXIO;
+	if (cc == 2)
+		return -ENXIO;
+	if (cc == 1)
+		return -EIO;
+	return 0;
+}
+
+/*
+ * Copy a 4kB page of main memory to an expanded memory page          
+ * Arguments                                                          
+ *           page_addr:    address of source page                     
+ *           xpage_index:  index of expandeded memory page            
+ * Return value                                                       
+ *           0:            if operation succeeds
+ *           -EIO:         if pgout failed
+ *           -ENXIO:       if xpram has vanished
+ */
+static long xpram_page_out (unsigned long page_addr, unsigned int xpage_index)
+{
+	int cc = 2;	/* return unused cc 2 if pgin traps */
+
+	asm volatile(
+		"	.insn	rre,0xb22f0000,%1,%2\n"  /* pgout %1,%2 */
+		"0:	ipm	%0\n"
+		"	srl	%0,28\n"
+		"1:\n"
+		EX_TABLE(0b,1b)
+		: "+d" (cc) : "a" (__pa(page_addr)), "d" (xpage_index) : "cc");
+	if (cc == 3)
+		return -ENXIO;
+	if (cc == 2)
+		return -ENXIO;
+	if (cc == 1)
+		return -EIO;
+	return 0;
+}
+
+/*
+ * Check if xpram is available.
+ */
+static int xpram_present(void)
+{
+	unsigned long mem_page;
+	int rc;
+
+	mem_page = (unsigned long) __get_free_page(GFP_KERNEL);
+	if (!mem_page)
+		return -ENOMEM;
+	rc = xpram_page_in(mem_page, 0);
+	free_page(mem_page);
+	return rc ? -ENXIO : 0;
+}
+
+/*
+ * Return index of the last available xpram page.
+ */
+static unsigned long xpram_highest_page_index(void)
+{
+	unsigned int page_index, add_bit;
+	unsigned long mem_page;
+
+	mem_page = (unsigned long) __get_free_page(GFP_KERNEL);
+	if (!mem_page)
+		return 0;
+
+	page_index = 0;
+	add_bit = 1ULL << (sizeof(unsigned int)*8 - 1);
+	while (add_bit > 0) {
+		if (xpram_page_in(mem_page, page_index | add_bit) == 0)
+			page_index |= add_bit;
+		add_bit >>= 1;
+	}
+
+	free_page (mem_page);
+
+	return page_index;
+}
+
+/*
+ * Block device make request function.
+ */
+static int xpram_make_request(struct request_queue *q, struct bio *bio)
+{
+	xpram_device_t *xdev = bio->bi_bdev->bd_disk->private_data;
+	struct bio_vec *bvec;
+	unsigned int index;
+	unsigned long page_addr;
+	unsigned long bytes;
+	int i;
+
+	if ((bio->bi_sector & 7) != 0 || (bio->bi_size & 4095) != 0)
+		/* Request is not page-aligned. */
+		goto fail;
+	if ((bio->bi_size >> 12) > xdev->size)
+		/* Request size is no page-aligned. */
+		goto fail;
+	if ((bio->bi_sector >> 3) > 0xffffffffU - xdev->offset)
+		goto fail;
+	index = (bio->bi_sector >> 3) + xdev->offset;
+	bio_for_each_segment(bvec, bio, i) {
+		page_addr = (unsigned long)
+			kmap(bvec->bv_page) + bvec->bv_offset;
+		bytes = bvec->bv_len;
+		if ((page_addr & 4095) != 0 || (bytes & 4095) != 0)
+			/* More paranoia. */
+			goto fail;
+		while (bytes > 0) {
+			if (bio_data_dir(bio) == READ) {
+				if (xpram_page_in(page_addr, index) != 0)
+					goto fail;
+			} else {
+				if (xpram_page_out(page_addr, index) != 0)
+					goto fail;
+			}
+			page_addr += 4096;
+			bytes -= 4096;
+			index++;
+		}
+	}
+	set_bit(BIO_UPTODATE, &bio->bi_flags);
+	bio_endio(bio, 0);
+	return 0;
+fail:
+	bio_io_error(bio);
+	return 0;
+}
+
+static int xpram_getgeo(struct block_device *bdev, struct hd_geometry *geo)
+{
+	unsigned long size;
+
+	/*
+	 * get geometry: we have to fake one...  trim the size to a
+	 * multiple of 64 (32k): tell we have 16 sectors, 4 heads,
+	 * whatever cylinders. Tell also that data starts at sector. 4.
+	 */
+	size = (xpram_pages * 8) & ~0x3f;
+	geo->cylinders = size >> 6;
+	geo->heads = 4;
+	geo->sectors = 16;
+	geo->start = 4;
+	return 0;
+}
+
+static const struct block_device_operations xpram_devops =
+{
+	.owner	= THIS_MODULE,
+	.getgeo	= xpram_getgeo,
+};
+
+/*
+ * Setup xpram_sizes array.
+ */
+static int __init xpram_setup_sizes(unsigned long pages)
+{
+	unsigned long mem_needed;
+	unsigned long mem_auto;
+	unsigned long long size;
+	int mem_auto_no;
+	int i;
+
+	/* Check number of devices. */
+	if (devs <= 0 || devs > XPRAM_MAX_DEVS) {
+		pr_err("%d is not a valid number of XPRAM devices\n",devs);
+		return -EINVAL;
+	}
+	xpram_devs = devs;
+
+	/*
+	 * Copy sizes array to xpram_sizes and align partition
+	 * sizes to page boundary.
+	 */
+	mem_needed = 0;
+	mem_auto_no = 0;
+	for (i = 0; i < xpram_devs; i++) {
+		if (sizes[i]) {
+			size = simple_strtoull(sizes[i], &sizes[i], 0);
+			switch (sizes[i][0]) {
+			case 'g':
+			case 'G':
+				size <<= 20;
+				break;
+			case 'm':
+			case 'M':
+				size <<= 10;
+			}
+			xpram_sizes[i] = (size + 3) & -4UL;
+		}
+		if (xpram_sizes[i])
+			mem_needed += xpram_sizes[i];
+		else
+			mem_auto_no++;
+	}
+	
+	pr_info("  number of devices (partitions): %d \n", xpram_devs);
+	for (i = 0; i < xpram_devs; i++) {
+		if (xpram_sizes[i])
+			pr_info("  size of partition %d: %u kB\n",
+				i, xpram_sizes[i]);
+		else
+			pr_info("  size of partition %d to be set "
+				"automatically\n",i);
+	}
+	pr_info("  memory needed (for sized partitions): %lu kB\n",
+		mem_needed);
+	pr_info("  partitions to be sized automatically: %d\n",
+		mem_auto_no);
+
+	if (mem_needed > pages * 4) {
+		pr_err("Not enough expanded memory available\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * partitioning:
+	 * xpram_sizes[i] != 0; partition i has size xpram_sizes[i] kB
+	 * else:             ; all partitions with zero xpram_sizes[i]
+	 *                     partition equally the remaining space
+	 */
+	if (mem_auto_no) {
+		mem_auto = ((pages - mem_needed / 4) / mem_auto_no) * 4;
+		pr_info("  automatically determined "
+			"partition size: %lu kB\n", mem_auto);
+		for (i = 0; i < xpram_devs; i++)
+			if (xpram_sizes[i] == 0)
+				xpram_sizes[i] = mem_auto;
+	}
+	return 0;
+}
+
+static int __init xpram_setup_blkdev(void)
+{
+	unsigned long offset;
+	int i, rc = -ENOMEM;
+
+	for (i = 0; i < xpram_devs; i++) {
+		xpram_disks[i] = alloc_disk(1);
+		if (!xpram_disks[i])
+			goto out;
+		xpram_queues[i] = blk_alloc_queue(GFP_KERNEL);
+		if (!xpram_queues[i]) {
+			put_disk(xpram_disks[i]);
+			goto out;
+		}
+		blk_queue_make_request(xpram_queues[i], xpram_make_request);
+		blk_queue_logical_block_size(xpram_queues[i], 4096);
+	}
+
+	/*
+	 * Register xpram major.
+	 */
+	rc = register_blkdev(XPRAM_MAJOR, XPRAM_NAME);
+	if (rc < 0)
+		goto out;
+
+	/*
+	 * Setup device structures.
+	 */
+	offset = 0;
+	for (i = 0; i < xpram_devs; i++) {
+		struct gendisk *disk = xpram_disks[i];
+
+		xpram_devices[i].size = xpram_sizes[i] / 4;
+		xpram_devices[i].offset = offset;
+		offset += xpram_devices[i].size;
+		disk->major = XPRAM_MAJOR;
+		disk->first_minor = i;
+		disk->fops = &xpram_devops;
+		disk->private_data = &xpram_devices[i];
+		disk->queue = xpram_queues[i];
+		sprintf(disk->disk_name, "slram%d", i);
+		set_capacity(disk, xpram_sizes[i] << 1);
+		add_disk(disk);
+	}
+
+	return 0;
+out:
+	while (i--) {
+		blk_cleanup_queue(xpram_queues[i]);
+		put_disk(xpram_disks[i]);
+	}
+	return rc;
+}
+
+/*
+ * Resume failed: Print error message and call panic.
+ */
+static void xpram_resume_error(const char *message)
+{
+	pr_err("Resuming the system failed: %s\n", message);
+	panic("xpram resume error\n");
+}
+
+/*
+ * Check if xpram setup changed between suspend and resume.
+ */
+static int xpram_restore(struct device *dev)
+{
+	if (!xpram_pages)
+		return 0;
+	if (xpram_present() != 0)
+		xpram_resume_error("xpram disappeared");
+	if (xpram_pages != xpram_highest_page_index() + 1)
+		xpram_resume_error("Size of xpram changed");
+	return 0;
+}
+
+static struct dev_pm_ops xpram_pm_ops = {
+	.restore	= xpram_restore,
+};
+
+static struct platform_driver xpram_pdrv = {
+	.driver = {
+		.name	= XPRAM_NAME,
+		.owner	= THIS_MODULE,
+		.pm	= &xpram_pm_ops,
+	},
+};
+
+static struct platform_device *xpram_pdev;
+
+/*
+ * Finally, the init/exit functions.
+ */
+static void __exit xpram_exit(void)
+{
+	int i;
+	for (i = 0; i < xpram_devs; i++) {
+		del_gendisk(xpram_disks[i]);
+		blk_cleanup_queue(xpram_queues[i]);
+		put_disk(xpram_disks[i]);
+	}
+	unregister_blkdev(XPRAM_MAJOR, XPRAM_NAME);
+	platform_device_unregister(xpram_pdev);
+	platform_driver_unregister(&xpram_pdrv);
+}
+
+static int __init xpram_init(void)
+{
+	int rc;
+
+	/* Find out size of expanded memory. */
+	if (xpram_present() != 0) {
+		pr_err("No expanded memory available\n");
+		return -ENODEV;
+	}
+	xpram_pages = xpram_highest_page_index() + 1;
+	pr_info("  %u pages expanded memory found (%lu KB).\n",
+		xpram_pages, (unsigned long) xpram_pages*4);
+	rc = xpram_setup_sizes(xpram_pages);
+	if (rc)
+		return rc;
+	rc = platform_driver_register(&xpram_pdrv);
+	if (rc)
+		return rc;
+	xpram_pdev = platform_device_register_simple(XPRAM_NAME, -1, NULL, 0);
+	if (IS_ERR(xpram_pdev)) {
+		rc = PTR_ERR(xpram_pdev);
+		goto fail_platform_driver_unregister;
+	}
+	rc = xpram_setup_blkdev();
+	if (rc)
+		goto fail_platform_device_unregister;
+	return 0;
+
+fail_platform_device_unregister:
+	platform_device_unregister(xpram_pdev);
+fail_platform_driver_unregister:
+	platform_driver_unregister(&xpram_pdrv);
+	return rc;
+}
+
+module_init(xpram_init);
+module_exit(xpram_exit);

kernel/drivers/s390/char/Kconfig

@@ -0,0 +1,177 @@
+comment "S/390 character device drivers"
+	depends on S390
+
+config TN3270
+	tristate "Support for locally attached 3270 terminals"
+	depends on CCW
+	help
+	  Include support for IBM 3270 terminals.
+
+config TN3270_TTY
+	tristate "Support for tty input/output on 3270 terminals"
+	depends on TN3270
+	help
+	  Include support for using an IBM 3270 terminal as a Linux tty.
+
+config TN3270_FS
+	tristate "Support for fullscreen applications on 3270 terminals"
+	depends on TN3270
+	help
+	  Include support for fullscreen applications on an IBM 3270 terminal.
+
+config TN3270_CONSOLE
+	bool "Support for console on 3270 terminal"
+	depends on TN3270=y && TN3270_TTY=y
+	help
+	  Include support for using an IBM 3270 terminal as a Linux system
+	  console.  Available only if 3270 support is compiled in statically.
+
+config TN3215
+	bool "Support for 3215 line mode terminal"
+	depends on CCW
+	help
+	  Include support for IBM 3215 line-mode terminals.
+
+config TN3215_CONSOLE
+	bool "Support for console on 3215 line mode terminal"
+	depends on TN3215
+	help
+	  Include support for using an IBM 3215 line-mode terminal as a
+	  Linux system console.
+
+config CCW_CONSOLE
+	bool
+	depends on TN3215_CONSOLE || TN3270_CONSOLE
+	default y
+
+config SCLP_TTY
+	bool "Support for SCLP line mode terminal"
+	depends on S390
+	help
+	  Include support for IBM SCLP line-mode terminals.
+
+config SCLP_CONSOLE
+	bool "Support for console on SCLP line mode terminal"
+	depends on SCLP_TTY
+	help
+	  Include support for using an IBM HWC line-mode terminal as the Linux
+	  system console.
+
+config SCLP_VT220_TTY
+	bool "Support for SCLP VT220-compatible terminal"
+	depends on S390
+	help
+	  Include support for an IBM SCLP VT220-compatible terminal.
+
+config SCLP_VT220_CONSOLE
+	bool "Support for console on SCLP VT220-compatible terminal"
+	depends on SCLP_VT220_TTY
+	help
+	  Include support for using an IBM SCLP VT220-compatible terminal as a
+	  Linux system console.
+
+config SCLP_CPI
+	tristate "Control-Program Identification"
+	depends on S390
+	help
+	  This option enables the hardware console interface for system
+	  identification. This is commonly used for workload management and
+	  gives you a nice name for the system on the service element.
+	  Please select this option as a module since built-in operation is
+	  completely untested.
+	  You should only select this option if you know what you are doing,
+	  need this feature and intend to run your kernel in LPAR.
+
+config SCLP_ASYNC
+	tristate "Support for Call Home via Asynchronous SCLP Records"
+	depends on S390
+	help
+	  This option enables the call home function, which is able to inform
+	  the service element and connected organisations about a kernel panic.
+	  You should only select this option if you know what you are doing,
+	  want for inform other people about your kernel panics,
+	  need this feature and intend to run your kernel in LPAR.
+
+config S390_TAPE
+	tristate "S/390 tape device support"
+	depends on CCW
+	help
+	  Select this option if you want to access channel-attached tape
+	  devices on IBM S/390 or zSeries.
+	  If you select this option you will also want to select at
+	  least one of the tape interface options and one of the tape
+	  hardware options in order to access a tape device.
+	  This option is also available as a module. The module will be
+	  called tape390 and include all selected interfaces and
+	  hardware drivers.
+
+comment "S/390 tape interface support"
+	depends on S390_TAPE
+
+config S390_TAPE_BLOCK
+	bool "Support for tape block devices"
+	depends on S390_TAPE && BLOCK
+	help
+	  Select this option if you want to access your channel-attached tape
+	  devices using the block device interface.  This interface is similar
+	  to CD-ROM devices on other platforms.  The tapes can only be
+	  accessed read-only when using this interface.  Have a look at
+	  <file:Documentation/s390/TAPE> for further information about creating
+	  volumes for and using this interface.  It is safe to say "Y" here.
+
+comment "S/390 tape hardware support"
+	depends on S390_TAPE
+
+config S390_TAPE_34XX
+	tristate "Support for 3480/3490 tape hardware"
+	depends on S390_TAPE
+	help
+	  Select this option if you want to access IBM 3480/3490 magnetic
+	  tape subsystems and 100% compatibles.
+	  It is safe to say "Y" here.
+
+config S390_TAPE_3590
+	tristate "Support for 3590 tape hardware"
+	depends on S390_TAPE
+	help
+	  Select this option if you want to access IBM 3590 magnetic
+	  tape subsystems and 100% compatibles.
+	  It is safe to say "Y" here.
+
+config VMLOGRDR
+	tristate "Support for the z/VM recording system services (VM only)"
+	depends on IUCV
+	help
+	  Select this option if you want to be able to receive records collected
+	  by the z/VM recording system services, eg. from *LOGREC, *ACCOUNT or
+	  *SYMPTOM.
+	  This driver depends on the IUCV support driver.
+
+config VMCP
+	tristate "Support for the z/VM CP interface (VM only)"
+	depends on S390
+	help
+	  Select this option if you want to be able to interact with the control
+	  program on z/VM
+
+
+config MONREADER
+	tristate "API for reading z/VM monitor service records"
+	depends on IUCV
+	help
+	  Character device driver for reading z/VM monitor service records
+
+config MONWRITER
+	tristate "API for writing z/VM monitor service records"
+	depends on S390
+	default "m"
+	help
+	  Character device driver for writing z/VM monitor service records
+
+config S390_VMUR
+	tristate "z/VM unit record device driver"
+	depends on S390
+	default "m"
+	help
+	  Character device driver for z/VM reader, puncher and printer.
+

kernel/drivers/s390/char/Makefile

@@ -0,0 +1,36 @@
+#
+# S/390 character devices
+#
+
+obj-y += ctrlchar.o keyboard.o defkeymap.o sclp.o sclp_rw.o sclp_quiesce.o \
+	 sclp_cmd.o sclp_config.o sclp_cpi_sys.o
+
+obj-$(CONFIG_TN3270) += raw3270.o
+obj-$(CONFIG_TN3270_CONSOLE) += con3270.o
+obj-$(CONFIG_TN3270_TTY) += tty3270.o
+obj-$(CONFIG_TN3270_FS) += fs3270.o
+
+obj-$(CONFIG_TN3215) += con3215.o
+
+obj-$(CONFIG_SCLP_TTY) += sclp_tty.o
+obj-$(CONFIG_SCLP_CONSOLE) += sclp_con.o
+obj-$(CONFIG_SCLP_VT220_TTY) += sclp_vt220.o
+obj-$(CONFIG_SCLP_CPI) += sclp_cpi.o
+obj-$(CONFIG_SCLP_ASYNC) += sclp_async.o
+
+obj-$(CONFIG_ZVM_WATCHDOG) += vmwatchdog.o
+obj-$(CONFIG_VMLOGRDR) += vmlogrdr.o
+obj-$(CONFIG_VMCP) += vmcp.o
+
+tape-$(CONFIG_S390_TAPE_BLOCK) += tape_block.o
+tape-$(CONFIG_PROC_FS) += tape_proc.o
+tape-objs := tape_core.o tape_std.o tape_char.o $(tape-y)
+obj-$(CONFIG_S390_TAPE) += tape.o tape_class.o
+obj-$(CONFIG_S390_TAPE_34XX) += tape_34xx.o
+obj-$(CONFIG_S390_TAPE_3590) += tape_3590.o
+obj-$(CONFIG_MONREADER) += monreader.o
+obj-$(CONFIG_MONWRITER) += monwriter.o
+obj-$(CONFIG_S390_VMUR) += vmur.o
+
+zcore_mod-objs := sclp_sdias.o zcore.o
+obj-$(CONFIG_ZFCPDUMP) += zcore_mod.o

kernel/drivers/s390/char/con3270.c

@@ -0,0 +1,636 @@
+/*
+ * IBM/3270 Driver - console view.
+ *
+ * Author(s):
+ *   Original 3270 Code for 2.4 written by Richard Hitt (UTS Global)
+ *   Rewritten for 2.5 by Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *     Copyright IBM Corp. 2003, 2009
+ */
+
+#include <linux/console.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/types.h>
+#include <linux/err.h>
+#include <linux/reboot.h>
+
+#include <asm/ccwdev.h>
+#include <asm/cio.h>
+#include <asm/cpcmd.h>
+#include <asm/ebcdic.h>
+
+#include "raw3270.h"
+#include "tty3270.h"
+#include "ctrlchar.h"
+
+#define CON3270_OUTPUT_BUFFER_SIZE 1024
+#define CON3270_STRING_PAGES 4
+
+static struct raw3270_fn con3270_fn;
+
+/*
+ * Main 3270 console view data structure.
+ */
+struct con3270 {
+	struct raw3270_view view;
+	spinlock_t lock;
+	struct list_head freemem;	/* list of free memory for strings. */
+
+	/* Output stuff. */
+	struct list_head lines;		/* list of lines. */
+	struct list_head update;	/* list of lines to update. */
+	int line_nr;			/* line number for next update. */
+	int nr_lines;			/* # lines in list. */
+	int nr_up;			/* # lines up in history. */
+	unsigned long update_flags;	/* Update indication bits. */
+	struct string *cline;		/* current output line. */
+	struct string *status;		/* last line of display. */
+	struct raw3270_request *write;	/* single write request. */
+	struct timer_list timer;
+
+	/* Input stuff. */
+	struct string *input;		/* input string for read request. */
+	struct raw3270_request *read;	/* single read request. */
+	struct raw3270_request *kreset;	/* single keyboard reset request. */
+	struct tasklet_struct readlet;	/* tasklet to issue read request. */
+};
+
+static struct con3270 *condev;
+
+/* con3270->update_flags. See con3270_update for details. */
+#define CON_UPDATE_ERASE	1	/* Use EWRITEA instead of WRITE. */
+#define CON_UPDATE_LIST		2	/* Update lines in tty3270->update. */
+#define CON_UPDATE_STATUS	4	/* Update status line. */
+#define CON_UPDATE_ALL		8	/* Recreate screen. */
+
+static void con3270_update(struct con3270 *);
+
+/*
+ * Setup timeout for a device. On timeout trigger an update.
+ */
+static void con3270_set_timer(struct con3270 *cp, int expires)
+{
+	if (expires == 0)
+		del_timer(&cp->timer);
+	else
+		mod_timer(&cp->timer, jiffies + expires);
+}
+
+/*
+ * The status line is the last line of the screen. It shows the string
+ * "console view" in the lower left corner and "Running"/"More..."/"Holding"
+ * in the lower right corner of the screen.
+ */
+static void
+con3270_update_status(struct con3270 *cp)
+{
+	char *str;
+
+	str = (cp->nr_up != 0) ? "History" : "Running";
+	memcpy(cp->status->string + 24, str, 7);
+	codepage_convert(cp->view.ascebc, cp->status->string + 24, 7);
+	cp->update_flags |= CON_UPDATE_STATUS;
+}
+
+static void
+con3270_create_status(struct con3270 *cp)
+{
+	static const unsigned char blueprint[] =
+		{ TO_SBA, 0, 0, TO_SF,TF_LOG,TO_SA,TAT_COLOR, TAC_GREEN,
+		  'c','o','n','s','o','l','e',' ','v','i','e','w',
+		  TO_RA,0,0,0,'R','u','n','n','i','n','g',TO_SF,TF_LOG };
+
+	cp->status = alloc_string(&cp->freemem, sizeof(blueprint));
+	/* Copy blueprint to status line */
+	memcpy(cp->status->string, blueprint, sizeof(blueprint));
+	/* Set TO_RA addresses. */
+	raw3270_buffer_address(cp->view.dev, cp->status->string + 1,
+			       cp->view.cols * (cp->view.rows - 1));
+	raw3270_buffer_address(cp->view.dev, cp->status->string + 21,
+			       cp->view.cols * cp->view.rows - 8);
+	/* Convert strings to ebcdic. */
+	codepage_convert(cp->view.ascebc, cp->status->string + 8, 12);
+	codepage_convert(cp->view.ascebc, cp->status->string + 24, 7);
+}
+
+/*
+ * Set output offsets to 3270 datastream fragment of a console string.
+ */
+static void
+con3270_update_string(struct con3270 *cp, struct string *s, int nr)
+{
+	if (s->len >= cp->view.cols - 5)
+		return;
+	raw3270_buffer_address(cp->view.dev, s->string + s->len - 3,
+			       cp->view.cols * (nr + 1));
+}
+
+/*
+ * Rebuild update list to print all lines.
+ */
+static void
+con3270_rebuild_update(struct con3270 *cp)
+{
+	struct string *s, *n;
+	int nr;
+
+	/* 
+	 * Throw away update list and create a new one,
+	 * containing all lines that will fit on the screen.
+	 */
+	list_for_each_entry_safe(s, n, &cp->update, update)
+		list_del_init(&s->update);
+	nr = cp->view.rows - 2 + cp->nr_up;
+	list_for_each_entry_reverse(s, &cp->lines, list) {
+		if (nr < cp->view.rows - 1)
+			list_add(&s->update, &cp->update);
+		if (--nr < 0)
+			break;
+	}
+	cp->line_nr = 0;
+	cp->update_flags |= CON_UPDATE_LIST;
+}
+
+/*
+ * Alloc string for size bytes. Free strings from history if necessary.
+ */
+static struct string *
+con3270_alloc_string(struct con3270 *cp, size_t size)
+{
+	struct string *s, *n;
+
+	s = alloc_string(&cp->freemem, size);
+	if (s)
+		return s;
+	list_for_each_entry_safe(s, n, &cp->lines, list) {
+		list_del(&s->list);
+		if (!list_empty(&s->update))
+			list_del(&s->update);
+		cp->nr_lines--;
+		if (free_string(&cp->freemem, s) >= size)
+			break;
+	}
+	s = alloc_string(&cp->freemem, size);
+	BUG_ON(!s);
+	if (cp->nr_up != 0 && cp->nr_up + cp->view.rows > cp->nr_lines) {
+		cp->nr_up = cp->nr_lines - cp->view.rows + 1;
+		con3270_rebuild_update(cp);
+		con3270_update_status(cp);
+	}
+	return s;
+}
+
+/*
+ * Write completion callback.
+ */
+static void
+con3270_write_callback(struct raw3270_request *rq, void *data)
+{
+	raw3270_request_reset(rq);
+	xchg(&((struct con3270 *) rq->view)->write, rq);
+}
+
+/*
+ * Update console display.
+ */
+static void
+con3270_update(struct con3270 *cp)
+{
+	struct raw3270_request *wrq;
+	char wcc, prolog[6];
+	unsigned long flags;
+	unsigned long updated;
+	struct string *s, *n;
+	int rc;
+
+	if (cp->view.dev)
+		raw3270_activate_view(&cp->view);
+
+	wrq = xchg(&cp->write, 0);
+	if (!wrq) {
+		con3270_set_timer(cp, 1);
+		return;
+	}
+
+	spin_lock_irqsave(&cp->view.lock, flags);
+	updated = 0;
+	if (cp->update_flags & CON_UPDATE_ALL) {
+		con3270_rebuild_update(cp);
+		con3270_update_status(cp);
+		cp->update_flags = CON_UPDATE_ERASE | CON_UPDATE_LIST |
+			CON_UPDATE_STATUS;
+	}
+	if (cp->update_flags & CON_UPDATE_ERASE) {
+		/* Use erase write alternate to initialize display. */
+		raw3270_request_set_cmd(wrq, TC_EWRITEA);
+		updated |= CON_UPDATE_ERASE;
+	} else
+		raw3270_request_set_cmd(wrq, TC_WRITE);
+
+	wcc = TW_NONE;
+	raw3270_request_add_data(wrq, &wcc, 1);
+
+	/*
+	 * Update status line.
+	 */
+	if (cp->update_flags & CON_UPDATE_STATUS)
+		if (raw3270_request_add_data(wrq, cp->status->string,
+					     cp->status->len) == 0)
+			updated |= CON_UPDATE_STATUS;
+
+	if (cp->update_flags & CON_UPDATE_LIST) {
+		prolog[0] = TO_SBA;
+		prolog[3] = TO_SA;
+		prolog[4] = TAT_COLOR;
+		prolog[5] = TAC_TURQ;
+		raw3270_buffer_address(cp->view.dev, prolog + 1,
+				       cp->view.cols * cp->line_nr);
+		raw3270_request_add_data(wrq, prolog, 6);
+		/* Write strings in the update list to the screen. */
+		list_for_each_entry_safe(s, n, &cp->update, update) {
+			if (s != cp->cline)
+				con3270_update_string(cp, s, cp->line_nr);
+			if (raw3270_request_add_data(wrq, s->string,
+						     s->len) != 0)
+				break;
+			list_del_init(&s->update);
+			if (s != cp->cline)
+				cp->line_nr++;
+		}
+		if (list_empty(&cp->update))
+			updated |= CON_UPDATE_LIST;
+	}
+	wrq->callback = con3270_write_callback;
+	rc = raw3270_start(&cp->view, wrq);
+	if (rc == 0) {
+		cp->update_flags &= ~updated;
+		if (cp->update_flags)
+			con3270_set_timer(cp, 1);
+	} else {
+		raw3270_request_reset(wrq);
+		xchg(&cp->write, wrq);
+	}
+	spin_unlock_irqrestore(&cp->view.lock, flags);
+}
+
+/*
+ * Read tasklet.
+ */
+static void
+con3270_read_tasklet(struct raw3270_request *rrq)
+{
+	static char kreset_data = TW_KR;
+	struct con3270 *cp;
+	unsigned long flags;
+	int nr_up, deactivate;
+
+	cp = (struct con3270 *) rrq->view;
+	spin_lock_irqsave(&cp->view.lock, flags);
+	nr_up = cp->nr_up;
+	deactivate = 0;
+	/* Check aid byte. */
+	switch (cp->input->string[0]) {
+	case 0x7d:	/* enter: jump to bottom. */
+		nr_up = 0;
+		break;
+	case 0xf3:	/* PF3: deactivate the console view. */
+		deactivate = 1;
+		break;
+	case 0x6d:	/* clear: start from scratch. */
+		cp->update_flags = CON_UPDATE_ALL;
+		con3270_set_timer(cp, 1);
+		break;
+	case 0xf7:	/* PF7: do a page up in the console log. */
+		nr_up += cp->view.rows - 2;
+		if (nr_up + cp->view.rows - 1 > cp->nr_lines) {
+			nr_up = cp->nr_lines - cp->view.rows + 1;
+			if (nr_up < 0)
+				nr_up = 0;
+		}
+		break;
+	case 0xf8:	/* PF8: do a page down in the console log. */
+		nr_up -= cp->view.rows - 2;
+		if (nr_up < 0)
+			nr_up = 0;
+		break;
+	}
+	if (nr_up != cp->nr_up) {
+		cp->nr_up = nr_up;
+		con3270_rebuild_update(cp);
+		con3270_update_status(cp);
+		con3270_set_timer(cp, 1);
+	}
+	spin_unlock_irqrestore(&cp->view.lock, flags);
+
+	/* Start keyboard reset command. */
+	raw3270_request_reset(cp->kreset);
+	raw3270_request_set_cmd(cp->kreset, TC_WRITE);
+	raw3270_request_add_data(cp->kreset, &kreset_data, 1);
+	raw3270_start(&cp->view, cp->kreset);
+
+	if (deactivate)
+		raw3270_deactivate_view(&cp->view);
+
+	raw3270_request_reset(rrq);
+	xchg(&cp->read, rrq);
+	raw3270_put_view(&cp->view);
+}
+
+/*
+ * Read request completion callback.
+ */
+static void
+con3270_read_callback(struct raw3270_request *rq, void *data)
+{
+	raw3270_get_view(rq->view);
+	/* Schedule tasklet to pass input to tty. */
+	tasklet_schedule(&((struct con3270 *) rq->view)->readlet);
+}
+
+/*
+ * Issue a read request. Called only from interrupt function.
+ */
+static void
+con3270_issue_read(struct con3270 *cp)
+{
+	struct raw3270_request *rrq;
+	int rc;
+
+	rrq = xchg(&cp->read, 0);
+	if (!rrq)
+		/* Read already scheduled. */
+		return;
+	rrq->callback = con3270_read_callback;
+	rrq->callback_data = cp;
+	raw3270_request_set_cmd(rrq, TC_READMOD);
+	raw3270_request_set_data(rrq, cp->input->string, cp->input->len);
+	/* Issue the read modified request. */
+	rc = raw3270_start_irq(&cp->view, rrq);
+	if (rc)
+		raw3270_request_reset(rrq);
+}
+
+/*
+ * Switch to the console view.
+ */
+static int
+con3270_activate(struct raw3270_view *view)
+{
+	struct con3270 *cp;
+
+	cp = (struct con3270 *) view;
+	cp->update_flags = CON_UPDATE_ALL;
+	con3270_set_timer(cp, 1);
+	return 0;
+}
+
+static void
+con3270_deactivate(struct raw3270_view *view)
+{
+	struct con3270 *cp;
+
+	cp = (struct con3270 *) view;
+	del_timer(&cp->timer);
+}
+
+static int
+con3270_irq(struct con3270 *cp, struct raw3270_request *rq, struct irb *irb)
+{
+	/* Handle ATTN. Schedule tasklet to read aid. */
+	if (irb->scsw.cmd.dstat & DEV_STAT_ATTENTION)
+		con3270_issue_read(cp);
+
+	if (rq) {
+		if (irb->scsw.cmd.dstat & DEV_STAT_UNIT_CHECK)
+			rq->rc = -EIO;
+		else
+			/* Normal end. Copy residual count. */
+			rq->rescnt = irb->scsw.cmd.count;
+	}
+	return RAW3270_IO_DONE;
+}
+
+/* Console view to a 3270 device. */
+static struct raw3270_fn con3270_fn = {
+	.activate = con3270_activate,
+	.deactivate = con3270_deactivate,
+	.intv = (void *) con3270_irq
+};
+
+static inline void
+con3270_cline_add(struct con3270 *cp)
+{
+	if (!list_empty(&cp->cline->list))
+		/* Already added. */
+		return;
+	list_add_tail(&cp->cline->list, &cp->lines);
+	cp->nr_lines++;
+	con3270_rebuild_update(cp);
+}
+
+static inline void
+con3270_cline_insert(struct con3270 *cp, unsigned char c)
+{
+	cp->cline->string[cp->cline->len++] = 
+		cp->view.ascebc[(c < ' ') ? ' ' : c];
+	if (list_empty(&cp->cline->update)) {
+		list_add_tail(&cp->cline->update, &cp->update);
+		cp->update_flags |= CON_UPDATE_LIST;
+	}
+}
+
+static inline void
+con3270_cline_end(struct con3270 *cp)
+{
+	struct string *s;
+	unsigned int size;
+
+	/* Copy cline. */
+	size = (cp->cline->len < cp->view.cols - 5) ?
+		cp->cline->len + 4 : cp->view.cols;
+	s = con3270_alloc_string(cp, size);
+	memcpy(s->string, cp->cline->string, cp->cline->len);
+	if (s->len < cp->view.cols - 5) {
+		s->string[s->len - 4] = TO_RA;
+		s->string[s->len - 1] = 0;
+	} else {
+		while (--size > cp->cline->len)
+			s->string[size] = cp->view.ascebc[' '];
+	}
+	/* Replace cline with allocated line s and reset cline. */
+	list_add(&s->list, &cp->cline->list);
+	list_del_init(&cp->cline->list);
+	if (!list_empty(&cp->cline->update)) {
+		list_add(&s->update, &cp->cline->update);
+		list_del_init(&cp->cline->update);
+	}
+	cp->cline->len = 0;
+}
+
+/*
+ * Write a string to the 3270 console
+ */
+static void
+con3270_write(struct console *co, const char *str, unsigned int count)
+{
+	struct con3270 *cp;
+	unsigned long flags;
+	unsigned char c;
+
+	cp = condev;
+	spin_lock_irqsave(&cp->view.lock, flags);
+	while (count-- > 0) {
+		c = *str++;
+		if (cp->cline->len == 0)
+			con3270_cline_add(cp);
+		if (c != '\n')
+			con3270_cline_insert(cp, c);
+		if (c == '\n' || cp->cline->len >= cp->view.cols)
+			con3270_cline_end(cp);
+	}
+	/* Setup timer to output current console buffer after 1/10 second */
+	cp->nr_up = 0;
+	if (cp->view.dev && !timer_pending(&cp->timer))
+		con3270_set_timer(cp, HZ/10);
+	spin_unlock_irqrestore(&cp->view.lock,flags);
+}
+
+static struct tty_driver *
+con3270_device(struct console *c, int *index)
+{
+	*index = c->index;
+	return tty3270_driver;
+}
+
+/*
+ * Wait for end of write request.
+ */
+static void
+con3270_wait_write(struct con3270 *cp)
+{
+	while (!cp->write) {
+		raw3270_wait_cons_dev(cp->view.dev);
+		barrier();
+	}
+}
+
+/*
+ * panic() calls con3270_flush through a panic_notifier
+ * before the system enters a disabled, endless loop.
+ */
+static void
+con3270_flush(void)
+{
+	struct con3270 *cp;
+	unsigned long flags;
+
+	cp = condev;
+	if (!cp->view.dev)
+		return;
+	raw3270_pm_unfreeze(&cp->view);
+	spin_lock_irqsave(&cp->view.lock, flags);
+	con3270_wait_write(cp);
+	cp->nr_up = 0;
+	con3270_rebuild_update(cp);
+	con3270_update_status(cp);
+	while (cp->update_flags != 0) {
+		spin_unlock_irqrestore(&cp->view.lock, flags);
+		con3270_update(cp);
+		spin_lock_irqsave(&cp->view.lock, flags);
+		con3270_wait_write(cp);
+	}
+	spin_unlock_irqrestore(&cp->view.lock, flags);
+}
+
+static int con3270_notify(struct notifier_block *self,
+			  unsigned long event, void *data)
+{
+	con3270_flush();
+	return NOTIFY_OK;
+}
+
+static struct notifier_block on_panic_nb = {
+	.notifier_call = con3270_notify,
+	.priority = 0,
+};
+
+static struct notifier_block on_reboot_nb = {
+	.notifier_call = con3270_notify,
+	.priority = 0,
+};
+
+/*
+ *  The console structure for the 3270 console
+ */
+static struct console con3270 = {
+	.name	 = "tty3270",
+	.write	 = con3270_write,
+	.device	 = con3270_device,
+	.flags	 = CON_PRINTBUFFER,
+};
+
+/*
+ * 3270 console initialization code called from console_init().
+ * NOTE: This is called before kmalloc is available.
+ */
+static int __init
+con3270_init(void)
+{
+	struct ccw_device *cdev;
+	struct raw3270 *rp;
+	void *cbuf;
+	int i;
+
+	/* Check if 3270 is to be the console */
+	if (!CONSOLE_IS_3270)
+		return -ENODEV;
+
+	/* Set the console mode for VM */
+	if (MACHINE_IS_VM) {
+		cpcmd("TERM CONMODE 3270", NULL, 0, NULL);
+		cpcmd("TERM AUTOCR OFF", NULL, 0, NULL);
+	}
+
+	cdev = ccw_device_probe_console();
+	if (IS_ERR(cdev))
+		return -ENODEV;
+	rp = raw3270_setup_console(cdev);
+	if (IS_ERR(rp))
+		return PTR_ERR(rp);
+
+	condev = kzalloc(sizeof(struct con3270), GFP_KERNEL | GFP_DMA);
+	condev->view.dev = rp;
+
+	condev->read = raw3270_request_alloc(0);
+	condev->read->callback = con3270_read_callback;
+	condev->read->callback_data = condev;
+	condev->write = raw3270_request_alloc(CON3270_OUTPUT_BUFFER_SIZE);
+	condev->kreset = raw3270_request_alloc(1);
+
+	INIT_LIST_HEAD(&condev->lines);
+	INIT_LIST_HEAD(&condev->update);
+	setup_timer(&condev->timer, (void (*)(unsigned long)) con3270_update,
+		    (unsigned long) condev);
+	tasklet_init(&condev->readlet, 
+		     (void (*)(unsigned long)) con3270_read_tasklet,
+		     (unsigned long) condev->read);
+
+	raw3270_add_view(&condev->view, &con3270_fn, 1);
+
+	INIT_LIST_HEAD(&condev->freemem);
+	for (i = 0; i < CON3270_STRING_PAGES; i++) {
+		cbuf = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		add_string_memory(&condev->freemem, cbuf, PAGE_SIZE);
+	}
+	condev->cline = alloc_string(&condev->freemem, condev->view.cols);
+	condev->cline->len = 0;
+	con3270_create_status(condev);
+	condev->input = alloc_string(&condev->freemem, 80);
+	atomic_notifier_chain_register(&panic_notifier_list, &on_panic_nb);
+	register_reboot_notifier(&on_reboot_nb);
+	register_console(&con3270);
+	return 0;
+}
+
+console_initcall(con3270_init);

kernel/drivers/s390/char/ctrlchar.c

@@ -0,0 +1,75 @@
+/*
+ *  drivers/s390/char/ctrlchar.c
+ *  Unified handling of special chars.
+ *
+ *    Copyright (C) 2001 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Author(s): Fritz Elfert <felfert@millenux.com> <elfert@de.ibm.com>
+ *
+ */
+
+#include <linux/stddef.h>
+#include <asm/errno.h>
+#include <linux/sysrq.h>
+#include <linux/ctype.h>
+
+#include "ctrlchar.h"
+
+#ifdef CONFIG_MAGIC_SYSRQ
+static int ctrlchar_sysrq_key;
+static struct tty_struct *sysrq_tty;
+
+static void
+ctrlchar_handle_sysrq(struct work_struct *work)
+{
+	handle_sysrq(ctrlchar_sysrq_key, sysrq_tty);
+}
+
+static DECLARE_WORK(ctrlchar_work, ctrlchar_handle_sysrq);
+#endif
+
+
+/**
+ * Check for special chars at start of input.
+ *
+ * @param buf Console input buffer.
+ * @param len Length of valid data in buffer.
+ * @param tty The tty struct for this console.
+ * @return CTRLCHAR_NONE, if nothing matched,
+ *         CTRLCHAR_SYSRQ, if sysrq was encountered
+ *         otherwise char to be inserted logically or'ed
+ *         with CTRLCHAR_CTRL
+ */
+unsigned int
+ctrlchar_handle(const unsigned char *buf, int len, struct tty_struct *tty)
+{
+	if ((len < 2) || (len > 3))
+		return CTRLCHAR_NONE;
+
+	/* hat is 0xb1 in codepage 037 (US etc.) and thus */
+	/* converted to 0x5e in ascii ('^') */
+	if ((buf[0] != '^') && (buf[0] != '\252'))
+		return CTRLCHAR_NONE;
+
+#ifdef CONFIG_MAGIC_SYSRQ
+	/* racy */
+	if (len == 3 && buf[1] == '-') {
+		ctrlchar_sysrq_key = buf[2];
+		sysrq_tty = tty;
+		schedule_work(&ctrlchar_work);
+		return CTRLCHAR_SYSRQ;
+	}
+#endif
+
+	if (len != 2)
+		return CTRLCHAR_NONE;
+
+	switch (tolower(buf[1])) {
+	case 'c':
+		return INTR_CHAR(tty) | CTRLCHAR_CTRL;
+	case 'd':
+		return EOF_CHAR(tty)  | CTRLCHAR_CTRL;
+	case 'z':
+		return SUSP_CHAR(tty) | CTRLCHAR_CTRL;
+	}
+	return CTRLCHAR_NONE;
+}

kernel/drivers/s390/char/ctrlchar.h

@@ -0,0 +1,20 @@
+/*
+ *  drivers/s390/char/ctrlchar.c
+ *  Unified handling of special chars.
+ *
+ *    Copyright (C) 2001 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Author(s): Fritz Elfert <felfert@millenux.com> <elfert@de.ibm.com>
+ *
+ */
+
+#include <linux/tty.h>
+
+extern unsigned int
+ctrlchar_handle(const unsigned char *buf, int len, struct tty_struct *tty);
+
+
+#define CTRLCHAR_NONE  (1 << 8)
+#define CTRLCHAR_CTRL  (2 << 8)
+#define CTRLCHAR_SYSRQ (3 << 8)
+
+#define CTRLCHAR_MASK (~0xffu)

kernel/drivers/s390/char/defkeymap.c

@@ -0,0 +1,158 @@
+
+/* Do not edit this file! It was automatically generated by   */
+/*    loadkeys --mktable defkeymap.map > defkeymap.c          */
+
+#include <linux/types.h>
+#include <linux/keyboard.h>
+#include <linux/kd.h>
+#include <linux/kbd_kern.h>
+#include <linux/kbd_diacr.h>
+
+u_short plain_map[NR_KEYS] = {
+	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,
+	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,
+	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,
+	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,
+	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,
+	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,
+	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,
+	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,	0xf000,
+	0xf020,	0xf000,	0xf0e2,	0xf0e4,	0xf0e0,	0xf0e1,	0xf0e3,	0xf0e5,
+	0xf0e7,	0xf0f1,	0xf0a2,	0xf02e,	0xf03c,	0xf028,	0xf02b,	0xf07c,
+	0xf026,	0xf0e9,	0xf0e2,	0xf0eb,	0xf0e8,	0xf0ed,	0xf0ee,	0xf0ef,
+	0xf0ec,	0xf0df,	0xf021,	0xf024,	0xf02a,	0xf029,	0xf03b,	0xf0ac,
+	0xf02d,	0xf02f,	0xf0c2,	0xf0c4,	0xf0c0,	0xf0c1,	0xf0c3,	0xf0c5,
+	0xf0c7,	0xf0d1,	0xf0a6,	0xf02c,	0xf025,	0xf05f,	0xf03e,	0xf03f,
+	0xf0f8,	0xf0c9,	0xf0ca,	0xf0cb,	0xf0c8,	0xf0cd,	0xf0ce,	0xf0cf,
+	0xf0cc,	0xf060,	0xf03a,	0xf023,	0xf040,	0xf027,	0xf03d,	0xf022,
+};
+
+static u_short shift_map[NR_KEYS] = {
+	0xf0d8,	0xf061,	0xf062,	0xf063,	0xf064,	0xf065,	0xf066,	0xf067,
+	0xf068,	0xf069,	0xf0ab,	0xf0bb,	0xf0f0,	0xf0fd,	0xf0fe,	0xf0b1,
+	0xf0b0,	0xf06a,	0xf06b,	0xf06c,	0xf06d,	0xf06e,	0xf06f,	0xf070,
+	0xf071,	0xf072,	0xf000,	0xf000,	0xf0e6,	0xf0b8,	0xf0c6,	0xf0a4,
+	0xf0b5,	0xf07e,	0xf073,	0xf074,	0xf075,	0xf076,	0xf077,	0xf078,
+	0xf079,	0xf07a,	0xf0a1,	0xf0bf,	0xf0d0,	0xf0dd,	0xf0de,	0xf0ae,
+	0xf402,	0xf0a3,	0xf0a5,	0xf0b7,	0xf0a9,	0xf0a7,	0xf0b6,	0xf0bc,
+	0xf0bd,	0xf0be,	0xf05b,	0xf05d,	0xf000,	0xf0a8,	0xf0b4,	0xf0d7,
+	0xf07b,	0xf041,	0xf042,	0xf043,	0xf044,	0xf045,	0xf046,	0xf047,
+	0xf048,	0xf049,	0xf000,	0xf0f4,	0xf0f6,	0xf0f2,	0xf0f3,	0xf0f5,
+	0xf07d,	0xf04a,	0xf04b,	0xf04c,	0xf04d,	0xf04e,	0xf04f,	0xf050,
+	0xf051,	0xf052,	0xf0b9,	0xf0fb,	0xf0fc,	0xf0f9,	0xf0fa,	0xf0ff,
+	0xf05c,	0xf0f7,	0xf053,	0xf054,	0xf055,	0xf056,	0xf057,	0xf058,
+	0xf059,	0xf05a,	0xf0b2,	0xf0d4,	0xf0d6,	0xf0d2,	0xf0d3,	0xf0d5,
+	0xf030,	0xf031,	0xf032,	0xf033,	0xf034,	0xf035,	0xf036,	0xf037,
+	0xf038,	0xf039,	0xf0b3,	0xf0db,	0xf0dc,	0xf0d9,	0xf0da,	0xf000,
+};
+
+static u_short ctrl_map[NR_KEYS] = {
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf11f,	0xf120,	0xf121,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf01a,	0xf003,	0xf212,	0xf004,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf109,	0xf10a,	0xf206,	0xf00a,	0xf200,	0xf200,
+};
+
+static u_short shift_ctrl_map[NR_KEYS] = {
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf10c,	0xf10d,	0xf10e,	0xf10f,	0xf110,	0xf111,	0xf112,
+	0xf113,	0xf11e,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+	0xf200,	0xf100,	0xf101,	0xf211,	0xf103,	0xf104,	0xf105,	0xf20b,
+	0xf20a,	0xf108,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,	0xf200,
+};
+
+ushort *key_maps[MAX_NR_KEYMAPS] = {
+	plain_map, shift_map, NULL, NULL,
+	ctrl_map, shift_ctrl_map, NULL,
+};
+
+unsigned int keymap_count = 4;
+
+
+/*
+ * Philosophy: most people do not define more strings, but they who do
+ * often want quite a lot of string space. So, we statically allocate
+ * the default and allocate dynamically in chunks of 512 bytes.
+ */
+
+char func_buf[] = {
+	'\033', '[', '[', 'A', 0, 
+	'\033', '[', '[', 'B', 0, 
+	'\033', '[', '[', 'C', 0, 
+	'\033', '[', '[', 'D', 0, 
+	'\033', '[', '[', 'E', 0, 
+	'\033', '[', '1', '7', '~', 0, 
+	'\033', '[', '1', '8', '~', 0, 
+	'\033', '[', '1', '9', '~', 0, 
+	'\033', '[', '2', '0', '~', 0, 
+	'\033', '[', '2', '1', '~', 0, 
+	'\033', '[', '2', '3', '~', 0, 
+	'\033', '[', '2', '4', '~', 0, 
+	'\033', '[', '2', '5', '~', 0, 
+	'\033', '[', '2', '6', '~', 0, 
+	'\033', '[', '2', '8', '~', 0, 
+	'\033', '[', '2', '9', '~', 0, 
+	'\033', '[', '3', '1', '~', 0, 
+	'\033', '[', '3', '2', '~', 0, 
+	'\033', '[', '3', '3', '~', 0, 
+	'\033', '[', '3', '4', '~', 0, 
+};
+
+
+char *funcbufptr = func_buf;
+int funcbufsize = sizeof(func_buf);
+int funcbufleft = 0;          /* space left */
+
+char *func_table[MAX_NR_FUNC] = {
+	func_buf + 0,
+	func_buf + 5,
+	func_buf + 10,
+	func_buf + 15,
+	func_buf + 20,
+	func_buf + 25,
+	func_buf + 31,
+	func_buf + 37,
+	func_buf + 43,
+	func_buf + 49,
+	func_buf + 55,
+	func_buf + 61,
+	func_buf + 67,
+	func_buf + 73,
+	func_buf + 79,
+	func_buf + 85,
+	func_buf + 91,
+	func_buf + 97,
+	func_buf + 103,
+	func_buf + 109,
+	NULL,
+};
+
+struct kbdiacruc accent_table[MAX_DIACR] = {
+	{'^', 'c', 0003},	{'^', 'd', 0004},
+	{'^', 'z', 0032},	{'^', 0012, 0000},
+};
+
+unsigned int accent_table_size = 4;

kernel/drivers/s390/char/defkeymap.map

@@ -0,0 +1,191 @@
+# Default keymap for 3270 (ebcdic codepage 037).
+keymaps 0-1,4-5
+
+keycode   0 = nul		Oslash
+keycode   1 = nul		a
+keycode   2 = nul		b
+keycode   3 = nul		c
+keycode   4 = nul		d
+keycode   5 = nul		e
+keycode   6 = nul		f
+keycode   7 = nul		g
+keycode   8 = nul		h
+keycode   9 = nul		i
+keycode  10 = nul		guillemotleft
+keycode  11 = nul		guillemotright
+keycode  12 = nul		eth
+keycode  13 = nul		yacute
+keycode  14 = nul		thorn
+keycode  15 = nul		plusminus
+keycode  16 = nul		degree
+keycode  17 = nul		j
+keycode  18 = nul		k
+keycode  19 = nul		l
+keycode  20 = nul		m
+keycode  21 = nul		n
+keycode  22 = nul		o
+keycode  23 = nul		p
+keycode  24 = nul		q
+keycode  25 = nul		r
+keycode  26 = nul		nul
+keycode  27 = nul		nul
+keycode  28 = nul		ae
+keycode  29 = nul		cedilla
+keycode  30 = nul		AE
+keycode  31 = nul		currency
+keycode  32 = nul		mu
+keycode  33 = nul		tilde
+keycode  34 = nul		s
+keycode  35 = nul		t
+keycode  36 = nul		u
+keycode  37 = nul		v
+keycode  38 = nul		w
+keycode  39 = nul		x
+keycode  40 = nul		y
+keycode  41 = nul		z
+keycode  42 = nul		exclamdown
+keycode  43 = nul		questiondown
+keycode  44 = nul		ETH
+keycode  45 = nul		Yacute
+keycode  46 = nul		THORN
+keycode  47 = nul		registered
+keycode  48 = nul		dead_circumflex
+keycode  49 = nul		sterling
+keycode  50 = nul		yen
+keycode  51 = nul		periodcentered
+keycode  52 = nul		copyright
+keycode  53 = nul		section
+keycode  54 = nul		paragraph
+keycode  55 = nul		onequarter
+keycode  56 = nul		onehalf
+keycode  57 = nul		threequarters
+keycode  58 = nul		bracketleft
+keycode  59 = nul		bracketright
+keycode  60 = nul		nul
+keycode  61 = nul		diaeresis
+keycode  62 = nul		acute
+keycode  63 = nul		multiply
+keycode  64 = space		braceleft
+keycode  65 = nul		A
+keycode  66 = acircumflex	B
+keycode  67 = adiaeresis	C
+keycode  68 = agrave		D
+keycode  69 = aacute		E
+keycode  70 = atilde		F
+keycode  71 = aring		G
+keycode  72 = ccedilla		H
+keycode  73 = ntilde		I
+keycode  74 = cent		nul
+keycode  75 = period		ocircumflex
+keycode  76 = less		odiaeresis
+keycode  77 = parenleft		ograve
+keycode  78 = plus		oacute
+keycode  79 = bar		otilde
+keycode  80 = ampersand		braceright
+keycode  81 = eacute		J
+keycode  82 = acircumflex	K
+keycode  83 = ediaeresis	L
+keycode  84 = egrave		M
+keycode  85 = iacute		N
+keycode  86 = icircumflex	O
+keycode  87 = idiaeresis	P
+keycode  88 = igrave		Q
+keycode  89 = ssharp		R
+keycode  90 = exclam		onesuperior
+keycode  91 = dollar		ucircumflex
+keycode  92 = asterisk		udiaeresis
+keycode  93 = parenright	ugrave
+keycode  94 = semicolon		uacute
+keycode  95 = notsign		ydiaeresis
+keycode  96 = minus		backslash
+keycode  97 = slash		division
+keycode  98 = Acircumflex	S
+keycode  99 = Adiaeresis	T
+keycode 100 = Agrave		U
+keycode 101 = Aacute		V
+keycode 102 = Atilde		W
+keycode 103 = Aring		X
+keycode 104 = Ccedilla		Y
+keycode 105 = Ntilde		Z
+keycode 106 = brokenbar		twosuperior
+keycode 107 = comma		Ocircumflex
+keycode 108 = percent		Odiaeresis
+keycode 109 = underscore	Ograve
+keycode 110 = greater		Oacute
+keycode 111 = question		Otilde
+keycode 112 = oslash		zero
+keycode 113 = Eacute		one
+keycode 114 = Ecircumflex	two
+keycode 115 = Ediaeresis	three
+keycode 116 = Egrave		four
+keycode 117 = Iacute		five
+keycode 118 = Icircumflex	six
+keycode 119 = Idiaeresis	seven
+keycode 120 = Igrave		eight
+keycode 121 = grave		nine
+keycode 122 = colon		threesuperior
+keycode 123 = numbersign	Ucircumflex
+keycode 124 = at		Udiaeresis
+keycode 125 = apostrophe	Ugrave
+keycode 126 = equal		Uacute
+keycode 127 = quotedbl		nul
+
+# AID keys
+control keycode  74 = F22
+control keycode  75 = F23
+control keycode  76 = F24
+control keycode 107 = Control_z		# PA3
+control keycode 108 = Control_c		# PA1
+control keycode 109 = KeyboardSignal	# Clear
+control keycode 110 = Control_d		# PA2
+control keycode 122 = F10
+control keycode 123 = F11		# F11
+control keycode 124 = Last_Console	# F12
+control keycode 125 = Linefeed
+shift control keycode  65 = F13
+shift control keycode  66 = F14
+shift control keycode  67 = F15
+shift control keycode  68 = F16
+shift control keycode  69 = F17
+shift control keycode  70 = F18
+shift control keycode  71 = F19
+shift control keycode  72 = F20
+shift control keycode  73 = F21
+shift control keycode 113 = F1
+shift control keycode 114 = F2
+shift control keycode 115 = Incr_Console
+shift control keycode 116 = F4
+shift control keycode 117 = F5
+shift control keycode 118 = F6
+shift control keycode 119 = Scroll_Backward
+shift control keycode 120 = Scroll_Forward
+shift control keycode 121 = F9
+
+string F1 = "\033[[A"
+string F2 = "\033[[B"
+string F3 = "\033[[C"
+string F4 = "\033[[D"
+string F5 = "\033[[E"
+string F6 = "\033[17~"
+string F7 = "\033[18~"
+string F8 = "\033[19~"
+string F9 = "\033[20~"
+string F10 = "\033[21~"
+string F11 = "\033[23~"
+string F12 = "\033[24~"
+string F13 = "\033[25~"
+string F14 = "\033[26~"
+string F15 = "\033[28~"
+string F16 = "\033[29~"
+string F17 = "\033[31~"
+string F18 = "\033[32~"
+string F19 = "\033[33~"
+string F20 = "\033[34~"
+# string F21 ??
+# string F22 ??
+# string F23 ??
+# string F24 ??
+compose '^' 'c' to Control_c
+compose '^' 'd' to Control_d
+compose '^' 'z' to Control_z
+compose '^' '\012' to nul

kernel/drivers/s390/char/fs3270.c

@@ -0,0 +1,540 @@
+/*
+ * IBM/3270 Driver - fullscreen driver.
+ *
+ * Author(s):
+ *   Original 3270 Code for 2.4 written by Richard Hitt (UTS Global)
+ *   Rewritten for 2.5/2.6 by Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *     Copyright IBM Corp. 2003, 2009
+ */
+
+#include <linux/bootmem.h>
+#include <linux/console.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/list.h>
+#include <linux/types.h>
+#include <linux/smp_lock.h>
+
+#include <asm/ccwdev.h>
+#include <asm/cio.h>
+#include <asm/ebcdic.h>
+#include <asm/idals.h>
+
+#include "raw3270.h"
+#include "ctrlchar.h"
+
+static struct raw3270_fn fs3270_fn;
+
+struct fs3270 {
+	struct raw3270_view view;
+	struct pid *fs_pid;		/* Pid of controlling program. */
+	int read_command;		/* ccw command to use for reads. */
+	int write_command;		/* ccw command to use for writes. */
+	int attention;			/* Got attention. */
+	int active;			/* Fullscreen view is active. */
+	struct raw3270_request *init;	/* single init request. */
+	wait_queue_head_t wait;		/* Init & attention wait queue. */
+	struct idal_buffer *rdbuf;	/* full-screen-deactivate buffer */
+	size_t rdbuf_size;		/* size of data returned by RDBUF */
+};
+
+static void
+fs3270_wake_up(struct raw3270_request *rq, void *data)
+{
+	wake_up((wait_queue_head_t *) data);
+}
+
+static inline int
+fs3270_working(struct fs3270 *fp)
+{
+	/*
+	 * The fullscreen view is in working order if the view
+	 * has been activated AND the initial request is finished.
+	 */
+	return fp->active && raw3270_request_final(fp->init);
+}
+
+static int
+fs3270_do_io(struct raw3270_view *view, struct raw3270_request *rq)
+{
+	struct fs3270 *fp;
+	int rc;
+
+	fp = (struct fs3270 *) view;
+	rq->callback = fs3270_wake_up;
+	rq->callback_data = &fp->wait;
+
+	do {
+		if (!fs3270_working(fp)) {
+			/* Fullscreen view isn't ready yet. */
+			rc = wait_event_interruptible(fp->wait,
+						      fs3270_working(fp));
+			if (rc != 0)
+				break;
+		}
+		rc = raw3270_start(view, rq);
+		if (rc == 0) {
+			/* Started sucessfully. Now wait for completion. */
+			wait_event(fp->wait, raw3270_request_final(rq));
+		}
+	} while (rc == -EACCES);
+	return rc;
+}
+
+/*
+ * Switch to the fullscreen view.
+ */
+static void
+fs3270_reset_callback(struct raw3270_request *rq, void *data)
+{
+	struct fs3270 *fp;
+
+	fp = (struct fs3270 *) rq->view;
+	raw3270_request_reset(rq);
+	wake_up(&fp->wait);
+}
+
+static void
+fs3270_restore_callback(struct raw3270_request *rq, void *data)
+{
+	struct fs3270 *fp;
+
+	fp = (struct fs3270 *) rq->view;
+	if (rq->rc != 0 || rq->rescnt != 0) {
+		if (fp->fs_pid)
+			kill_pid(fp->fs_pid, SIGHUP, 1);
+	}
+	fp->rdbuf_size = 0;
+	raw3270_request_reset(rq);
+	wake_up(&fp->wait);
+}
+
+static int
+fs3270_activate(struct raw3270_view *view)
+{
+	struct fs3270 *fp;
+	char *cp;
+	int rc;
+
+	fp = (struct fs3270 *) view;
+
+	/* If an old init command is still running just return. */
+	if (!raw3270_request_final(fp->init))
+		return 0;
+
+	if (fp->rdbuf_size == 0) {
+		/* No saved buffer. Just clear the screen. */
+		raw3270_request_set_cmd(fp->init, TC_EWRITEA);
+		fp->init->callback = fs3270_reset_callback;
+	} else {
+		/* Restore fullscreen buffer saved by fs3270_deactivate. */
+		raw3270_request_set_cmd(fp->init, TC_EWRITEA);
+		raw3270_request_set_idal(fp->init, fp->rdbuf);
+		fp->init->ccw.count = fp->rdbuf_size;
+		cp = fp->rdbuf->data[0];
+		cp[0] = TW_KR;
+		cp[1] = TO_SBA;
+		cp[2] = cp[6];
+		cp[3] = cp[7];
+		cp[4] = TO_IC;
+		cp[5] = TO_SBA;
+		cp[6] = 0x40;
+		cp[7] = 0x40;
+		fp->init->rescnt = 0;
+		fp->init->callback = fs3270_restore_callback;
+	}
+	rc = fp->init->rc = raw3270_start_locked(view, fp->init);
+	if (rc)
+		fp->init->callback(fp->init, NULL);
+	else
+		fp->active = 1;
+	return rc;
+}
+
+/*
+ * Shutdown fullscreen view.
+ */
+static void
+fs3270_save_callback(struct raw3270_request *rq, void *data)
+{
+	struct fs3270 *fp;
+
+	fp = (struct fs3270 *) rq->view;
+
+	/* Correct idal buffer element 0 address. */
+	fp->rdbuf->data[0] -= 5;
+	fp->rdbuf->size += 5;
+
+	/*
+	 * If the rdbuf command failed or the idal buffer is
+	 * to small for the amount of data returned by the
+	 * rdbuf command, then we have no choice but to send
+	 * a SIGHUP to the application.
+	 */
+	if (rq->rc != 0 || rq->rescnt == 0) {
+		if (fp->fs_pid)
+			kill_pid(fp->fs_pid, SIGHUP, 1);
+		fp->rdbuf_size = 0;
+	} else
+		fp->rdbuf_size = fp->rdbuf->size - rq->rescnt;
+	raw3270_request_reset(rq);
+	wake_up(&fp->wait);
+}
+
+static void
+fs3270_deactivate(struct raw3270_view *view)
+{
+	struct fs3270 *fp;
+
+	fp = (struct fs3270 *) view;
+	fp->active = 0;
+
+	/* If an old init command is still running just return. */
+	if (!raw3270_request_final(fp->init))
+		return;
+
+	/* Prepare read-buffer request. */
+	raw3270_request_set_cmd(fp->init, TC_RDBUF);
+	/*
+	 * Hackish: skip first 5 bytes of the idal buffer to make
+	 * room for the TW_KR/TO_SBA/<address>/<address>/TO_IC sequence
+	 * in the activation command.
+	 */
+	fp->rdbuf->data[0] += 5;
+	fp->rdbuf->size -= 5;
+	raw3270_request_set_idal(fp->init, fp->rdbuf);
+	fp->init->rescnt = 0;
+	fp->init->callback = fs3270_save_callback;
+
+	/* Start I/O to read in the 3270 buffer. */
+	fp->init->rc = raw3270_start_locked(view, fp->init);
+	if (fp->init->rc)
+		fp->init->callback(fp->init, NULL);
+}
+
+static int
+fs3270_irq(struct fs3270 *fp, struct raw3270_request *rq, struct irb *irb)
+{
+	/* Handle ATTN. Set indication and wake waiters for attention. */
+	if (irb->scsw.cmd.dstat & DEV_STAT_ATTENTION) {
+		fp->attention = 1;
+		wake_up(&fp->wait);
+	}
+
+	if (rq) {
+		if (irb->scsw.cmd.dstat & DEV_STAT_UNIT_CHECK)
+			rq->rc = -EIO;
+		else
+			/* Normal end. Copy residual count. */
+			rq->rescnt = irb->scsw.cmd.count;
+	}
+	return RAW3270_IO_DONE;
+}
+
+/*
+ * Process reads from fullscreen 3270.
+ */
+static ssize_t
+fs3270_read(struct file *filp, char __user *data, size_t count, loff_t *off)
+{
+	struct fs3270 *fp;
+	struct raw3270_request *rq;
+	struct idal_buffer *ib;
+	ssize_t rc;
+	
+	if (count == 0 || count > 65535)
+		return -EINVAL;
+	fp = filp->private_data;
+	if (!fp)
+		return -ENODEV;
+	ib = idal_buffer_alloc(count, 0);
+	if (IS_ERR(ib))
+		return -ENOMEM;
+	rq = raw3270_request_alloc(0);
+	if (!IS_ERR(rq)) {
+		if (fp->read_command == 0 && fp->write_command != 0)
+			fp->read_command = 6;
+		raw3270_request_set_cmd(rq, fp->read_command ? : 2);
+		raw3270_request_set_idal(rq, ib);
+		rc = wait_event_interruptible(fp->wait, fp->attention);
+		fp->attention = 0;
+		if (rc == 0) {
+			rc = fs3270_do_io(&fp->view, rq);
+			if (rc == 0) {
+				count -= rq->rescnt;
+				if (idal_buffer_to_user(ib, data, count) != 0)
+					rc = -EFAULT;
+				else
+					rc = count;
+
+			}
+		}
+		raw3270_request_free(rq);
+	} else
+		rc = PTR_ERR(rq);
+	idal_buffer_free(ib);
+	return rc;
+}
+
+/*
+ * Process writes to fullscreen 3270.
+ */
+static ssize_t
+fs3270_write(struct file *filp, const char __user *data, size_t count, loff_t *off)
+{
+	struct fs3270 *fp;
+	struct raw3270_request *rq;
+	struct idal_buffer *ib;
+	int write_command;
+	ssize_t rc;
+
+	fp = filp->private_data;
+	if (!fp)
+		return -ENODEV;
+	ib = idal_buffer_alloc(count, 0);
+	if (IS_ERR(ib))
+		return -ENOMEM;
+	rq = raw3270_request_alloc(0);
+	if (!IS_ERR(rq)) {
+		if (idal_buffer_from_user(ib, data, count) == 0) {
+			write_command = fp->write_command ? : 1;
+			if (write_command == 5)
+				write_command = 13;
+			raw3270_request_set_cmd(rq, write_command);
+			raw3270_request_set_idal(rq, ib);
+			rc = fs3270_do_io(&fp->view, rq);
+			if (rc == 0)
+				rc = count - rq->rescnt;
+		} else
+			rc = -EFAULT;
+		raw3270_request_free(rq);
+	} else
+		rc = PTR_ERR(rq);
+	idal_buffer_free(ib);
+	return rc;
+}
+
+/*
+ * process ioctl commands for the tube driver
+ */
+static long
+fs3270_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
+{
+	struct fs3270 *fp;
+	struct raw3270_iocb iocb;
+	int rc;
+
+	fp = filp->private_data;
+	if (!fp)
+		return -ENODEV;
+	rc = 0;
+	lock_kernel();
+	switch (cmd) {
+	case TUBICMD:
+		fp->read_command = arg;
+		break;
+	case TUBOCMD:
+		fp->write_command = arg;
+		break;
+	case TUBGETI:
+		rc = put_user(fp->read_command, (char __user *) arg);
+		break;
+	case TUBGETO:
+		rc = put_user(fp->write_command,(char __user *) arg);
+		break;
+	case TUBGETMOD:
+		iocb.model = fp->view.model;
+		iocb.line_cnt = fp->view.rows;
+		iocb.col_cnt = fp->view.cols;
+		iocb.pf_cnt = 24;
+		iocb.re_cnt = 20;
+		iocb.map = 0;
+		if (copy_to_user((char __user *) arg, &iocb,
+				 sizeof(struct raw3270_iocb)))
+			rc = -EFAULT;
+		break;
+	}
+	unlock_kernel();
+	return rc;
+}
+
+/*
+ * Allocate fs3270 structure.
+ */
+static struct fs3270 *
+fs3270_alloc_view(void)
+{
+	struct fs3270 *fp;
+
+	fp = kzalloc(sizeof(struct fs3270),GFP_KERNEL);
+	if (!fp)
+		return ERR_PTR(-ENOMEM);
+	fp->init = raw3270_request_alloc(0);
+	if (IS_ERR(fp->init)) {
+		kfree(fp);
+		return ERR_PTR(-ENOMEM);
+	}
+	return fp;
+}
+
+/*
+ * Free fs3270 structure.
+ */
+static void
+fs3270_free_view(struct raw3270_view *view)
+{
+	struct fs3270 *fp;
+
+	fp = (struct fs3270 *) view;
+	if (fp->rdbuf)
+		idal_buffer_free(fp->rdbuf);
+	raw3270_request_free(((struct fs3270 *) view)->init);
+	kfree(view);
+}
+
+/*
+ * Unlink fs3270 data structure from filp.
+ */
+static void
+fs3270_release(struct raw3270_view *view)
+{
+	struct fs3270 *fp;
+
+	fp = (struct fs3270 *) view;
+	if (fp->fs_pid)
+		kill_pid(fp->fs_pid, SIGHUP, 1);
+}
+
+/* View to a 3270 device. Can be console, tty or fullscreen. */
+static struct raw3270_fn fs3270_fn = {
+	.activate = fs3270_activate,
+	.deactivate = fs3270_deactivate,
+	.intv = (void *) fs3270_irq,
+	.release = fs3270_release,
+	.free = fs3270_free_view
+};
+
+/*
+ * This routine is called whenever a 3270 fullscreen device is opened.
+ */
+static int
+fs3270_open(struct inode *inode, struct file *filp)
+{
+	struct fs3270 *fp;
+	struct idal_buffer *ib;
+	int minor, rc = 0;
+
+	if (imajor(filp->f_path.dentry->d_inode) != IBM_FS3270_MAJOR)
+		return -ENODEV;
+	minor = iminor(filp->f_path.dentry->d_inode);
+	/* Check for minor 0 multiplexer. */
+	if (minor == 0) {
+		struct tty_struct *tty = get_current_tty();
+		if (!tty || tty->driver->major != IBM_TTY3270_MAJOR) {
+			tty_kref_put(tty);
+			return -ENODEV;
+		}
+		minor = tty->index + RAW3270_FIRSTMINOR;
+		tty_kref_put(tty);
+	}
+	lock_kernel();
+	/* Check if some other program is already using fullscreen mode. */
+	fp = (struct fs3270 *) raw3270_find_view(&fs3270_fn, minor);
+	if (!IS_ERR(fp)) {
+		raw3270_put_view(&fp->view);
+		rc = -EBUSY;
+		goto out;
+	}
+	/* Allocate fullscreen view structure. */
+	fp = fs3270_alloc_view();
+	if (IS_ERR(fp)) {
+		rc = PTR_ERR(fp);
+		goto out;
+	}
+
+	init_waitqueue_head(&fp->wait);
+	fp->fs_pid = get_pid(task_pid(current));
+	rc = raw3270_add_view(&fp->view, &fs3270_fn, minor);
+	if (rc) {
+		fs3270_free_view(&fp->view);
+		goto out;
+	}
+
+	/* Allocate idal-buffer. */
+	ib = idal_buffer_alloc(2*fp->view.rows*fp->view.cols + 5, 0);
+	if (IS_ERR(ib)) {
+		raw3270_put_view(&fp->view);
+		raw3270_del_view(&fp->view);
+		rc = PTR_ERR(fp);
+		goto out;
+	}
+	fp->rdbuf = ib;
+
+	rc = raw3270_activate_view(&fp->view);
+	if (rc) {
+		raw3270_put_view(&fp->view);
+		raw3270_del_view(&fp->view);
+		goto out;
+	}
+	filp->private_data = fp;
+out:
+	unlock_kernel();
+	return rc;
+}
+
+/*
+ * This routine is called when the 3270 tty is closed. We wait
+ * for the remaining request to be completed. Then we clean up.
+ */
+static int
+fs3270_close(struct inode *inode, struct file *filp)
+{
+	struct fs3270 *fp;
+
+	fp = filp->private_data;
+	filp->private_data = NULL;
+	if (fp) {
+		put_pid(fp->fs_pid);
+		fp->fs_pid = NULL;
+		raw3270_reset(&fp->view);
+		raw3270_put_view(&fp->view);
+		raw3270_del_view(&fp->view);
+	}
+	return 0;
+}
+
+static const struct file_operations fs3270_fops = {
+	.owner		 = THIS_MODULE,		/* owner */
+	.read		 = fs3270_read,		/* read */
+	.write		 = fs3270_write,	/* write */
+	.unlocked_ioctl	 = fs3270_ioctl,	/* ioctl */
+	.compat_ioctl	 = fs3270_ioctl,	/* ioctl */
+	.open	 	= fs3270_open,		/* open */
+	.release 	= fs3270_close,		/* release */
+};
+
+/*
+ * 3270 fullscreen driver initialization.
+ */
+static int __init
+fs3270_init(void)
+{
+	int rc;
+
+	rc = register_chrdev(IBM_FS3270_MAJOR, "fs3270", &fs3270_fops);
+	if (rc)
+		return rc;
+	return 0;
+}
+
+static void __exit
+fs3270_exit(void)
+{
+	unregister_chrdev(IBM_FS3270_MAJOR, "fs3270");
+}
+
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_CHARDEV_MAJOR(IBM_FS3270_MAJOR);
+
+module_init(fs3270_init);
+module_exit(fs3270_exit);

kernel/drivers/s390/char/keyboard.c

@@ -0,0 +1,560 @@
+/*
+ *  drivers/s390/char/keyboard.c
+ *    ebcdic keycode functions for s390 console drivers
+ *
+ *  S390 version
+ *    Copyright (C) 2003 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
+ */
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/sysrq.h>
+
+#include <linux/consolemap.h>
+#include <linux/kbd_kern.h>
+#include <linux/kbd_diacr.h>
+#include <asm/uaccess.h>
+
+#include "keyboard.h"
+
+/*
+ * Handler Tables.
+ */
+#define K_HANDLERS\
+	k_self,		k_fn,		k_spec,		k_ignore,\
+	k_dead,		k_ignore,	k_ignore,	k_ignore,\
+	k_ignore,	k_ignore,	k_ignore,	k_ignore,\
+	k_ignore,	k_ignore,	k_ignore,	k_ignore
+
+typedef void (k_handler_fn)(struct kbd_data *, unsigned char);
+static k_handler_fn K_HANDLERS;
+static k_handler_fn *k_handler[16] = { K_HANDLERS };
+
+/* maximum values each key_handler can handle */
+static const int kbd_max_vals[] = {
+	255, ARRAY_SIZE(func_table) - 1, NR_FN_HANDLER - 1, 0,
+	NR_DEAD - 1, 0, 0, 0, 0, 0, 0, 0, 0, 0
+};
+static const int KBD_NR_TYPES = ARRAY_SIZE(kbd_max_vals);
+
+static unsigned char ret_diacr[NR_DEAD] = {
+	'`', '\'', '^', '~', '"', ','
+};
+
+/*
+ * Alloc/free of kbd_data structures.
+ */
+struct kbd_data *
+kbd_alloc(void) {
+	struct kbd_data *kbd;
+	int i, len;
+
+	kbd = kzalloc(sizeof(struct kbd_data), GFP_KERNEL);
+	if (!kbd)
+		goto out;
+	kbd->key_maps = kzalloc(sizeof(key_maps), GFP_KERNEL);
+	if (!kbd->key_maps)
+		goto out_kbd;
+	for (i = 0; i < ARRAY_SIZE(key_maps); i++) {
+		if (key_maps[i]) {
+			kbd->key_maps[i] =
+				kmalloc(sizeof(u_short)*NR_KEYS, GFP_KERNEL);
+			if (!kbd->key_maps[i])
+				goto out_maps;
+			memcpy(kbd->key_maps[i], key_maps[i],
+			       sizeof(u_short)*NR_KEYS);
+		}
+	}
+	kbd->func_table = kzalloc(sizeof(func_table), GFP_KERNEL);
+	if (!kbd->func_table)
+		goto out_maps;
+	for (i = 0; i < ARRAY_SIZE(func_table); i++) {
+		if (func_table[i]) {
+			len = strlen(func_table[i]) + 1;
+			kbd->func_table[i] = kmalloc(len, GFP_KERNEL);
+			if (!kbd->func_table[i])
+				goto out_func;
+			memcpy(kbd->func_table[i], func_table[i], len);
+		}
+	}
+	kbd->fn_handler =
+		kzalloc(sizeof(fn_handler_fn *) * NR_FN_HANDLER, GFP_KERNEL);
+	if (!kbd->fn_handler)
+		goto out_func;
+	kbd->accent_table =
+		kmalloc(sizeof(struct kbdiacruc)*MAX_DIACR, GFP_KERNEL);
+	if (!kbd->accent_table)
+		goto out_fn_handler;
+	memcpy(kbd->accent_table, accent_table,
+	       sizeof(struct kbdiacruc)*MAX_DIACR);
+	kbd->accent_table_size = accent_table_size;
+	return kbd;
+
+out_fn_handler:
+	kfree(kbd->fn_handler);
+out_func:
+	for (i = 0; i < ARRAY_SIZE(func_table); i++)
+		kfree(kbd->func_table[i]);
+	kfree(kbd->func_table);
+out_maps:
+	for (i = 0; i < ARRAY_SIZE(key_maps); i++)
+		kfree(kbd->key_maps[i]);
+	kfree(kbd->key_maps);
+out_kbd:
+	kfree(kbd);
+out:
+	return NULL;
+}
+
+void
+kbd_free(struct kbd_data *kbd)
+{
+	int i;
+
+	kfree(kbd->accent_table);
+	kfree(kbd->fn_handler);
+	for (i = 0; i < ARRAY_SIZE(func_table); i++)
+		kfree(kbd->func_table[i]);
+	kfree(kbd->func_table);
+	for (i = 0; i < ARRAY_SIZE(key_maps); i++)
+		kfree(kbd->key_maps[i]);
+	kfree(kbd->key_maps);
+	kfree(kbd);
+}
+
+/*
+ * Generate ascii -> ebcdic translation table from kbd_data.
+ */
+void
+kbd_ascebc(struct kbd_data *kbd, unsigned char *ascebc)
+{
+	unsigned short *keymap, keysym;
+	int i, j, k;
+
+	memset(ascebc, 0x40, 256);
+	for (i = 0; i < ARRAY_SIZE(key_maps); i++) {
+		keymap = kbd->key_maps[i];
+		if (!keymap)
+			continue;
+		for (j = 0; j < NR_KEYS; j++) {
+			k = ((i & 1) << 7) + j;
+			keysym = keymap[j];
+			if (KTYP(keysym) == (KT_LATIN | 0xf0) ||
+			    KTYP(keysym) == (KT_LETTER | 0xf0))
+				ascebc[KVAL(keysym)] = k;
+			else if (KTYP(keysym) == (KT_DEAD | 0xf0))
+				ascebc[ret_diacr[KVAL(keysym)]] = k;
+		}
+	}
+}
+
+#if 0
+/*
+ * Generate ebcdic -> ascii translation table from kbd_data.
+ */
+void
+kbd_ebcasc(struct kbd_data *kbd, unsigned char *ebcasc)
+{
+	unsigned short *keymap, keysym;
+	int i, j, k;
+
+	memset(ebcasc, ' ', 256);
+	for (i = 0; i < ARRAY_SIZE(key_maps); i++) {
+		keymap = kbd->key_maps[i];
+		if (!keymap)
+			continue;
+		for (j = 0; j < NR_KEYS; j++) {
+			keysym = keymap[j];
+			k = ((i & 1) << 7) + j;
+			if (KTYP(keysym) == (KT_LATIN | 0xf0) ||
+			    KTYP(keysym) == (KT_LETTER | 0xf0))
+				ebcasc[k] = KVAL(keysym);
+			else if (KTYP(keysym) == (KT_DEAD | 0xf0))
+				ebcasc[k] = ret_diacr[KVAL(keysym)];
+		}
+	}
+}
+#endif
+
+/*
+ * We have a combining character DIACR here, followed by the character CH.
+ * If the combination occurs in the table, return the corresponding value.
+ * Otherwise, if CH is a space or equals DIACR, return DIACR.
+ * Otherwise, conclude that DIACR was not combining after all,
+ * queue it and return CH.
+ */
+static unsigned int
+handle_diacr(struct kbd_data *kbd, unsigned int ch)
+{
+	int i, d;
+
+	d = kbd->diacr;
+	kbd->diacr = 0;
+
+	for (i = 0; i < kbd->accent_table_size; i++) {
+		if (kbd->accent_table[i].diacr == d &&
+		    kbd->accent_table[i].base == ch)
+			return kbd->accent_table[i].result;
+	}
+
+	if (ch == ' ' || ch == d)
+		return d;
+
+	kbd_put_queue(kbd->tty, d);
+	return ch;
+}
+
+/*
+ * Handle dead key.
+ */
+static void
+k_dead(struct kbd_data *kbd, unsigned char value)
+{
+	value = ret_diacr[value];
+	kbd->diacr = (kbd->diacr ? handle_diacr(kbd, value) : value);
+}
+
+/*
+ * Normal character handler.
+ */
+static void
+k_self(struct kbd_data *kbd, unsigned char value)
+{
+	if (kbd->diacr)
+		value = handle_diacr(kbd, value);
+	kbd_put_queue(kbd->tty, value);
+}
+
+/*
+ * Special key handlers
+ */
+static void
+k_ignore(struct kbd_data *kbd, unsigned char value)
+{
+}
+
+/*
+ * Function key handler.
+ */
+static void
+k_fn(struct kbd_data *kbd, unsigned char value)
+{
+	if (kbd->func_table[value])
+		kbd_puts_queue(kbd->tty, kbd->func_table[value]);
+}
+
+static void
+k_spec(struct kbd_data *kbd, unsigned char value)
+{
+	if (value >= NR_FN_HANDLER)
+		return;
+	if (kbd->fn_handler[value])
+		kbd->fn_handler[value](kbd);
+}
+
+/*
+ * Put utf8 character to tty flip buffer.
+ * UTF-8 is defined for words of up to 31 bits,
+ * but we need only 16 bits here
+ */
+static void
+to_utf8(struct tty_struct *tty, ushort c) 
+{
+	if (c < 0x80)
+		/*  0******* */
+		kbd_put_queue(tty, c);
+	else if (c < 0x800) {
+		/* 110***** 10****** */
+		kbd_put_queue(tty, 0xc0 | (c >> 6));
+		kbd_put_queue(tty, 0x80 | (c & 0x3f));
+	} else {
+		/* 1110**** 10****** 10****** */
+		kbd_put_queue(tty, 0xe0 | (c >> 12));
+		kbd_put_queue(tty, 0x80 | ((c >> 6) & 0x3f));
+		kbd_put_queue(tty, 0x80 | (c & 0x3f));
+	}
+}
+
+/*
+ * Process keycode.
+ */
+void
+kbd_keycode(struct kbd_data *kbd, unsigned int keycode)
+{
+	unsigned short keysym;
+	unsigned char type, value;
+
+	if (!kbd || !kbd->tty)
+		return;
+
+	if (keycode >= 384)
+		keysym = kbd->key_maps[5][keycode - 384];
+	else if (keycode >= 256)
+		keysym = kbd->key_maps[4][keycode - 256];
+	else if (keycode >= 128)
+		keysym = kbd->key_maps[1][keycode - 128];
+	else
+		keysym = kbd->key_maps[0][keycode];
+
+	type = KTYP(keysym);
+	if (type >= 0xf0) {
+		type -= 0xf0;
+		if (type == KT_LETTER)
+			type = KT_LATIN;
+		value = KVAL(keysym);
+#ifdef CONFIG_MAGIC_SYSRQ	       /* Handle the SysRq Hack */
+		if (kbd->sysrq) {
+			if (kbd->sysrq == K(KT_LATIN, '-')) {
+				kbd->sysrq = 0;
+				handle_sysrq(value, kbd->tty);
+				return;
+			}
+			if (value == '-') {
+				kbd->sysrq = K(KT_LATIN, '-');
+				return;
+			}
+			/* Incomplete sysrq sequence. */
+			(*k_handler[KTYP(kbd->sysrq)])(kbd, KVAL(kbd->sysrq));
+			kbd->sysrq = 0;
+		} else if ((type == KT_LATIN && value == '^') ||
+			   (type == KT_DEAD && ret_diacr[value] == '^')) {
+			kbd->sysrq = K(type, value);
+			return;
+		}
+#endif
+		(*k_handler[type])(kbd, value);
+	} else
+		to_utf8(kbd->tty, keysym);
+}
+
+/*
+ * Ioctl stuff.
+ */
+static int
+do_kdsk_ioctl(struct kbd_data *kbd, struct kbentry __user *user_kbe,
+	      int cmd, int perm)
+{
+	struct kbentry tmp;
+	ushort *key_map, val, ov;
+
+	if (copy_from_user(&tmp, user_kbe, sizeof(struct kbentry)))
+		return -EFAULT;
+#if NR_KEYS < 256
+	if (tmp.kb_index >= NR_KEYS)
+		return -EINVAL;
+#endif
+#if MAX_NR_KEYMAPS < 256
+	if (tmp.kb_table >= MAX_NR_KEYMAPS)
+		return -EINVAL;	
+#endif
+
+	switch (cmd) {
+	case KDGKBENT:
+		key_map = kbd->key_maps[tmp.kb_table];
+		if (key_map) {
+		    val = U(key_map[tmp.kb_index]);
+		    if (KTYP(val) >= KBD_NR_TYPES)
+			val = K_HOLE;
+		} else
+		    val = (tmp.kb_index ? K_HOLE : K_NOSUCHMAP);
+		return put_user(val, &user_kbe->kb_value);
+	case KDSKBENT:
+		if (!perm)
+			return -EPERM;
+		if (!tmp.kb_index && tmp.kb_value == K_NOSUCHMAP) {
+			/* disallocate map */
+			key_map = kbd->key_maps[tmp.kb_table];
+			if (key_map) {
+			    kbd->key_maps[tmp.kb_table] = NULL;
+			    kfree(key_map);
+			}
+			break;
+		}
+
+		if (KTYP(tmp.kb_value) >= KBD_NR_TYPES)
+			return -EINVAL;
+		if (KVAL(tmp.kb_value) > kbd_max_vals[KTYP(tmp.kb_value)])
+			return -EINVAL;
+
+		if (!(key_map = kbd->key_maps[tmp.kb_table])) {
+			int j;
+
+			key_map = kmalloc(sizeof(plain_map),
+						     GFP_KERNEL);
+			if (!key_map)
+				return -ENOMEM;
+			kbd->key_maps[tmp.kb_table] = key_map;
+			for (j = 0; j < NR_KEYS; j++)
+				key_map[j] = U(K_HOLE);
+		}
+		ov = U(key_map[tmp.kb_index]);
+		if (tmp.kb_value == ov)
+			break;	/* nothing to do */
+		/*
+		 * Attention Key.
+		 */
+		if (((ov == K_SAK) || (tmp.kb_value == K_SAK)) &&
+		    !capable(CAP_SYS_ADMIN))
+			return -EPERM;
+		key_map[tmp.kb_index] = U(tmp.kb_value);
+		break;
+	}
+	return 0;
+}
+
+static int
+do_kdgkb_ioctl(struct kbd_data *kbd, struct kbsentry __user *u_kbs,
+	       int cmd, int perm)
+{
+	unsigned char kb_func;
+	char *p;
+	int len;
+
+	/* Get u_kbs->kb_func. */
+	if (get_user(kb_func, &u_kbs->kb_func))
+		return -EFAULT;
+#if MAX_NR_FUNC < 256
+	if (kb_func >= MAX_NR_FUNC)
+		return -EINVAL;
+#endif
+
+	switch (cmd) {
+	case KDGKBSENT:
+		p = kbd->func_table[kb_func];
+		if (p) {
+			len = strlen(p);
+			if (len >= sizeof(u_kbs->kb_string))
+				len = sizeof(u_kbs->kb_string) - 1;
+			if (copy_to_user(u_kbs->kb_string, p, len))
+				return -EFAULT;
+		} else
+			len = 0;
+		if (put_user('\0', u_kbs->kb_string + len))
+			return -EFAULT;
+		break;
+	case KDSKBSENT:
+		if (!perm)
+			return -EPERM;
+		len = strnlen_user(u_kbs->kb_string,
+				   sizeof(u_kbs->kb_string) - 1);
+		if (!len)
+			return -EFAULT;
+		if (len > sizeof(u_kbs->kb_string) - 1)
+			return -EINVAL;
+		p = kmalloc(len + 1, GFP_KERNEL);
+		if (!p)
+			return -ENOMEM;
+		if (copy_from_user(p, u_kbs->kb_string, len)) {
+			kfree(p);
+			return -EFAULT;
+		}
+		p[len] = 0;
+		kfree(kbd->func_table[kb_func]);
+		kbd->func_table[kb_func] = p;
+		break;
+	}
+	return 0;
+}
+
+int
+kbd_ioctl(struct kbd_data *kbd, struct file *file,
+	  unsigned int cmd, unsigned long arg)
+{
+	void __user *argp;
+	unsigned int ct;
+	int perm;
+
+	argp = (void __user *)arg;
+
+	/*
+	 * To have permissions to do most of the vt ioctls, we either have
+	 * to be the owner of the tty, or have CAP_SYS_TTY_CONFIG.
+	 */
+	perm = current->signal->tty == kbd->tty || capable(CAP_SYS_TTY_CONFIG);
+	switch (cmd) {
+	case KDGKBTYPE:
+		return put_user(KB_101, (char __user *)argp);
+	case KDGKBENT:
+	case KDSKBENT:
+		return do_kdsk_ioctl(kbd, argp, cmd, perm);
+	case KDGKBSENT:
+	case KDSKBSENT:
+		return do_kdgkb_ioctl(kbd, argp, cmd, perm);
+	case KDGKBDIACR:
+	{
+		struct kbdiacrs __user *a = argp;
+		struct kbdiacr diacr;
+		int i;
+
+		if (put_user(kbd->accent_table_size, &a->kb_cnt))
+			return -EFAULT;
+		for (i = 0; i < kbd->accent_table_size; i++) {
+			diacr.diacr = kbd->accent_table[i].diacr;
+			diacr.base = kbd->accent_table[i].base;
+			diacr.result = kbd->accent_table[i].result;
+			if (copy_to_user(a->kbdiacr + i, &diacr, sizeof(struct kbdiacr)))
+			return -EFAULT;
+		}
+		return 0;
+	}
+	case KDGKBDIACRUC:
+	{
+		struct kbdiacrsuc __user *a = argp;
+
+		ct = kbd->accent_table_size;
+		if (put_user(ct, &a->kb_cnt))
+			return -EFAULT;
+		if (copy_to_user(a->kbdiacruc, kbd->accent_table,
+				 ct * sizeof(struct kbdiacruc)))
+			return -EFAULT;
+		return 0;
+	}
+	case KDSKBDIACR:
+	{
+		struct kbdiacrs __user *a = argp;
+		struct kbdiacr diacr;
+		int i;
+
+		if (!perm)
+			return -EPERM;
+		if (get_user(ct, &a->kb_cnt))
+			return -EFAULT;
+		if (ct >= MAX_DIACR)
+			return -EINVAL;
+		kbd->accent_table_size = ct;
+		for (i = 0; i < ct; i++) {
+			if (copy_from_user(&diacr, a->kbdiacr + i, sizeof(struct kbdiacr)))
+				return -EFAULT;
+			kbd->accent_table[i].diacr = diacr.diacr;
+			kbd->accent_table[i].base = diacr.base;
+			kbd->accent_table[i].result = diacr.result;
+		}
+		return 0;
+	}
+	case KDSKBDIACRUC:
+	{
+		struct kbdiacrsuc __user *a = argp;
+
+		if (!perm)
+			return -EPERM;
+		if (get_user(ct, &a->kb_cnt))
+			return -EFAULT;
+		if (ct >= MAX_DIACR)
+			return -EINVAL;
+		kbd->accent_table_size = ct;
+		if (copy_from_user(kbd->accent_table, a->kbdiacruc,
+				   ct * sizeof(struct kbdiacruc)))
+			return -EFAULT;
+		return 0;
+	}
+	default:
+		return -ENOIOCTLCMD;
+	}
+}
+
+EXPORT_SYMBOL(kbd_ioctl);
+EXPORT_SYMBOL(kbd_ascebc);
+EXPORT_SYMBOL(kbd_free);
+EXPORT_SYMBOL(kbd_alloc);
+EXPORT_SYMBOL(kbd_keycode);

kernel/drivers/s390/char/keyboard.h

@@ -0,0 +1,57 @@
+/*
+ *  drivers/s390/char/keyboard.h
+ *    ebcdic keycode functions for s390 console drivers
+ *
+ *    Copyright (C) 2003 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
+ */
+
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include <linux/keyboard.h>
+
+#define NR_FN_HANDLER	20
+
+struct kbd_data;
+
+typedef void (fn_handler_fn)(struct kbd_data *);
+
+/*
+ * FIXME: explain key_maps tricks.
+ */
+
+struct kbd_data {
+	struct tty_struct *tty;
+	unsigned short **key_maps;
+	char **func_table;
+	fn_handler_fn **fn_handler;
+	struct kbdiacruc *accent_table;
+	unsigned int accent_table_size;
+	unsigned int diacr;
+	unsigned short sysrq;
+};
+
+struct kbd_data *kbd_alloc(void);
+void kbd_free(struct kbd_data *);
+void kbd_ascebc(struct kbd_data *, unsigned char *);
+
+void kbd_keycode(struct kbd_data *, unsigned int);
+int kbd_ioctl(struct kbd_data *, struct file *, unsigned int, unsigned long);
+
+/*
+ * Helper Functions.
+ */
+static inline void
+kbd_put_queue(struct tty_struct *tty, int ch)
+{
+	tty_insert_flip_char(tty, ch, 0);
+	tty_schedule_flip(tty);
+}
+
+static inline void
+kbd_puts_queue(struct tty_struct *tty, char *cp)
+{
+	while (*cp)
+		tty_insert_flip_char(tty, *cp++, 0);
+	tty_schedule_flip(tty);
+}

kernel/drivers/s390/char/monreader.c

@@ -0,0 +1,647 @@
+/*
+ * Character device driver for reading z/VM *MONITOR service records.
+ *
+ * Copyright IBM Corp. 2004, 2009
+ *
+ * Author: Gerald Schaefer <gerald.schaefer@de.ibm.com>
+ */
+
+#define KMSG_COMPONENT "monreader"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/init.h>
+#include <linux/smp_lock.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/ctype.h>
+#include <linux/spinlock.h>
+#include <linux/interrupt.h>
+#include <linux/poll.h>
+#include <linux/device.h>
+#include <net/iucv/iucv.h>
+#include <asm/uaccess.h>
+#include <asm/ebcdic.h>
+#include <asm/extmem.h>
+
+
+#define MON_COLLECT_SAMPLE 0x80
+#define MON_COLLECT_EVENT  0x40
+#define MON_SERVICE	   "*MONITOR"
+#define MON_IN_USE	   0x01
+#define MON_MSGLIM	   255
+
+static char mon_dcss_name[9] = "MONDCSS\0";
+
+struct mon_msg {
+	u32 pos;
+	u32 mca_offset;
+	struct iucv_message msg;
+	char msglim_reached;
+	char replied_msglim;
+};
+
+struct mon_private {
+	struct iucv_path *path;
+	struct mon_msg *msg_array[MON_MSGLIM];
+	unsigned int   write_index;
+	unsigned int   read_index;
+	atomic_t msglim_count;
+	atomic_t read_ready;
+	atomic_t iucv_connected;
+	atomic_t iucv_severed;
+};
+
+static unsigned long mon_in_use = 0;
+
+static unsigned long mon_dcss_start;
+static unsigned long mon_dcss_end;
+
+static DECLARE_WAIT_QUEUE_HEAD(mon_read_wait_queue);
+static DECLARE_WAIT_QUEUE_HEAD(mon_conn_wait_queue);
+
+static u8 user_data_connect[16] = {
+	/* Version code, must be 0x01 for shared mode */
+	0x01,
+	/* what to collect */
+	MON_COLLECT_SAMPLE | MON_COLLECT_EVENT,
+	/* DCSS name in EBCDIC, 8 bytes padded with blanks */
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+};
+
+static u8 user_data_sever[16] = {
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+	0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
+};
+
+static struct device *monreader_device;
+
+/******************************************************************************
+ *                             helper functions                               *
+ *****************************************************************************/
+/*
+ * Create the 8 bytes EBCDIC DCSS segment name from
+ * an ASCII name, incl. padding
+ */
+static void dcss_mkname(char *ascii_name, char *ebcdic_name)
+{
+	int i;
+
+	for (i = 0; i < 8; i++) {
+		if (ascii_name[i] == '\0')
+			break;
+		ebcdic_name[i] = toupper(ascii_name[i]);
+	};
+	for (; i < 8; i++)
+		ebcdic_name[i] = ' ';
+	ASCEBC(ebcdic_name, 8);
+}
+
+static inline unsigned long mon_mca_start(struct mon_msg *monmsg)
+{
+	return *(u32 *) &monmsg->msg.rmmsg;
+}
+
+static inline unsigned long mon_mca_end(struct mon_msg *monmsg)
+{
+	return *(u32 *) &monmsg->msg.rmmsg[4];
+}
+
+static inline u8 mon_mca_type(struct mon_msg *monmsg, u8 index)
+{
+	return *((u8 *) mon_mca_start(monmsg) + monmsg->mca_offset + index);
+}
+
+static inline u32 mon_mca_size(struct mon_msg *monmsg)
+{
+	return mon_mca_end(monmsg) - mon_mca_start(monmsg) + 1;
+}
+
+static inline u32 mon_rec_start(struct mon_msg *monmsg)
+{
+	return *((u32 *) (mon_mca_start(monmsg) + monmsg->mca_offset + 4));
+}
+
+static inline u32 mon_rec_end(struct mon_msg *monmsg)
+{
+	return *((u32 *) (mon_mca_start(monmsg) + monmsg->mca_offset + 8));
+}
+
+static int mon_check_mca(struct mon_msg *monmsg)
+{
+	if ((mon_rec_end(monmsg) <= mon_rec_start(monmsg)) ||
+	    (mon_rec_start(monmsg) < mon_dcss_start) ||
+	    (mon_rec_end(monmsg) > mon_dcss_end) ||
+	    (mon_mca_type(monmsg, 0) == 0) ||
+	    (mon_mca_size(monmsg) % 12 != 0) ||
+	    (mon_mca_end(monmsg) <= mon_mca_start(monmsg)) ||
+	    (mon_mca_end(monmsg) > mon_dcss_end) ||
+	    (mon_mca_start(monmsg) < mon_dcss_start) ||
+	    ((mon_mca_type(monmsg, 1) == 0) && (mon_mca_type(monmsg, 2) == 0)))
+		return -EINVAL;
+	return 0;
+}
+
+static int mon_send_reply(struct mon_msg *monmsg,
+			  struct mon_private *monpriv)
+{
+	int rc;
+
+	rc = iucv_message_reply(monpriv->path, &monmsg->msg,
+				IUCV_IPRMDATA, NULL, 0);
+	atomic_dec(&monpriv->msglim_count);
+	if (likely(!monmsg->msglim_reached)) {
+		monmsg->pos = 0;
+		monmsg->mca_offset = 0;
+		monpriv->read_index = (monpriv->read_index + 1) %
+				      MON_MSGLIM;
+		atomic_dec(&monpriv->read_ready);
+	} else
+		monmsg->replied_msglim = 1;
+	if (rc) {
+		pr_err("Reading monitor data failed with rc=%i\n", rc);
+		return -EIO;
+	}
+	return 0;
+}
+
+static void mon_free_mem(struct mon_private *monpriv)
+{
+	int i;
+
+	for (i = 0; i < MON_MSGLIM; i++)
+		if (monpriv->msg_array[i])
+			kfree(monpriv->msg_array[i]);
+	kfree(monpriv);
+}
+
+static struct mon_private *mon_alloc_mem(void)
+{
+	int i;
+	struct mon_private *monpriv;
+
+	monpriv = kzalloc(sizeof(struct mon_private), GFP_KERNEL);
+	if (!monpriv)
+		return NULL;
+	for (i = 0; i < MON_MSGLIM; i++) {
+		monpriv->msg_array[i] = kzalloc(sizeof(struct mon_msg),
+						    GFP_KERNEL);
+		if (!monpriv->msg_array[i]) {
+			mon_free_mem(monpriv);
+			return NULL;
+		}
+	}
+	return monpriv;
+}
+
+static inline void mon_next_mca(struct mon_msg *monmsg)
+{
+	if (likely((mon_mca_size(monmsg) - monmsg->mca_offset) == 12))
+		return;
+	monmsg->mca_offset += 12;
+	monmsg->pos = 0;
+}
+
+static struct mon_msg *mon_next_message(struct mon_private *monpriv)
+{
+	struct mon_msg *monmsg;
+
+	if (!atomic_read(&monpriv->read_ready))
+		return NULL;
+	monmsg = monpriv->msg_array[monpriv->read_index];
+	if (unlikely(monmsg->replied_msglim)) {
+		monmsg->replied_msglim = 0;
+		monmsg->msglim_reached = 0;
+		monmsg->pos = 0;
+		monmsg->mca_offset = 0;
+		monpriv->read_index = (monpriv->read_index + 1) %
+				      MON_MSGLIM;
+		atomic_dec(&monpriv->read_ready);
+		return ERR_PTR(-EOVERFLOW);
+	}
+	return monmsg;
+}
+
+
+/******************************************************************************
+ *                               IUCV handler                                 *
+ *****************************************************************************/
+static void mon_iucv_path_complete(struct iucv_path *path, u8 ipuser[16])
+{
+	struct mon_private *monpriv = path->private;
+
+	atomic_set(&monpriv->iucv_connected, 1);
+	wake_up(&mon_conn_wait_queue);
+}
+
+static void mon_iucv_path_severed(struct iucv_path *path, u8 ipuser[16])
+{
+	struct mon_private *monpriv = path->private;
+
+	pr_err("z/VM *MONITOR system service disconnected with rc=%i\n",
+	       ipuser[0]);
+	iucv_path_sever(path, NULL);
+	atomic_set(&monpriv->iucv_severed, 1);
+	wake_up(&mon_conn_wait_queue);
+	wake_up_interruptible(&mon_read_wait_queue);
+}
+
+static void mon_iucv_message_pending(struct iucv_path *path,
+				     struct iucv_message *msg)
+{
+	struct mon_private *monpriv = path->private;
+
+	memcpy(&monpriv->msg_array[monpriv->write_index]->msg,
+	       msg, sizeof(*msg));
+	if (atomic_inc_return(&monpriv->msglim_count) == MON_MSGLIM) {
+		pr_warning("The read queue for monitor data is full\n");
+		monpriv->msg_array[monpriv->write_index]->msglim_reached = 1;
+	}
+	monpriv->write_index = (monpriv->write_index + 1) % MON_MSGLIM;
+	atomic_inc(&monpriv->read_ready);
+	wake_up_interruptible(&mon_read_wait_queue);
+}
+
+static struct iucv_handler monreader_iucv_handler = {
+	.path_complete	 = mon_iucv_path_complete,
+	.path_severed	 = mon_iucv_path_severed,
+	.message_pending = mon_iucv_message_pending,
+};
+
+/******************************************************************************
+ *                               file operations                              *
+ *****************************************************************************/
+static int mon_open(struct inode *inode, struct file *filp)
+{
+	struct mon_private *monpriv;
+	int rc;
+
+	/*
+	 * only one user allowed
+	 */
+	lock_kernel();
+	rc = -EBUSY;
+	if (test_and_set_bit(MON_IN_USE, &mon_in_use))
+		goto out;
+
+	rc = -ENOMEM;
+	monpriv = mon_alloc_mem();
+	if (!monpriv)
+		goto out_use;
+
+	/*
+	 * Connect to *MONITOR service
+	 */
+	monpriv->path = iucv_path_alloc(MON_MSGLIM, IUCV_IPRMDATA, GFP_KERNEL);
+	if (!monpriv->path)
+		goto out_priv;
+	rc = iucv_path_connect(monpriv->path, &monreader_iucv_handler,
+			       MON_SERVICE, NULL, user_data_connect, monpriv);
+	if (rc) {
+		pr_err("Connecting to the z/VM *MONITOR system service "
+		       "failed with rc=%i\n", rc);
+		rc = -EIO;
+		goto out_path;
+	}
+	/*
+	 * Wait for connection confirmation
+	 */
+	wait_event(mon_conn_wait_queue,
+		   atomic_read(&monpriv->iucv_connected) ||
+		   atomic_read(&monpriv->iucv_severed));
+	if (atomic_read(&monpriv->iucv_severed)) {
+		atomic_set(&monpriv->iucv_severed, 0);
+		atomic_set(&monpriv->iucv_connected, 0);
+		rc = -EIO;
+		goto out_path;
+	}
+	filp->private_data = monpriv;
+	dev_set_drvdata(monreader_device, monpriv);
+	unlock_kernel();
+	return nonseekable_open(inode, filp);
+
+out_path:
+	iucv_path_free(monpriv->path);
+out_priv:
+	mon_free_mem(monpriv);
+out_use:
+	clear_bit(MON_IN_USE, &mon_in_use);
+out:
+	unlock_kernel();
+	return rc;
+}
+
+static int mon_close(struct inode *inode, struct file *filp)
+{
+	int rc, i;
+	struct mon_private *monpriv = filp->private_data;
+
+	/*
+	 * Close IUCV connection and unregister
+	 */
+	if (monpriv->path) {
+		rc = iucv_path_sever(monpriv->path, user_data_sever);
+		if (rc)
+			pr_warning("Disconnecting the z/VM *MONITOR system "
+				   "service failed with rc=%i\n", rc);
+		iucv_path_free(monpriv->path);
+	}
+
+	atomic_set(&monpriv->iucv_severed, 0);
+	atomic_set(&monpriv->iucv_connected, 0);
+	atomic_set(&monpriv->read_ready, 0);
+	atomic_set(&monpriv->msglim_count, 0);
+	monpriv->write_index  = 0;
+	monpriv->read_index   = 0;
+	dev_set_drvdata(monreader_device, NULL);
+
+	for (i = 0; i < MON_MSGLIM; i++)
+		kfree(monpriv->msg_array[i]);
+	kfree(monpriv);
+	clear_bit(MON_IN_USE, &mon_in_use);
+	return 0;
+}
+
+static ssize_t mon_read(struct file *filp, char __user *data,
+			size_t count, loff_t *ppos)
+{
+	struct mon_private *monpriv = filp->private_data;
+	struct mon_msg *monmsg;
+	int ret;
+	u32 mce_start;
+
+	monmsg = mon_next_message(monpriv);
+	if (IS_ERR(monmsg))
+		return PTR_ERR(monmsg);
+
+	if (!monmsg) {
+		if (filp->f_flags & O_NONBLOCK)
+			return -EAGAIN;
+		ret = wait_event_interruptible(mon_read_wait_queue,
+					atomic_read(&monpriv->read_ready) ||
+					atomic_read(&monpriv->iucv_severed));
+		if (ret)
+			return ret;
+		if (unlikely(atomic_read(&monpriv->iucv_severed)))
+			return -EIO;
+		monmsg = monpriv->msg_array[monpriv->read_index];
+	}
+
+	if (!monmsg->pos)
+		monmsg->pos = mon_mca_start(monmsg) + monmsg->mca_offset;
+	if (mon_check_mca(monmsg))
+		goto reply;
+
+	/* read monitor control element (12 bytes) first */
+	mce_start = mon_mca_start(monmsg) + monmsg->mca_offset;
+	if ((monmsg->pos >= mce_start) && (monmsg->pos < mce_start + 12)) {
+		count = min(count, (size_t) mce_start + 12 - monmsg->pos);
+		ret = copy_to_user(data, (void *) (unsigned long) monmsg->pos,
+				   count);
+		if (ret)
+			return -EFAULT;
+		monmsg->pos += count;
+		if (monmsg->pos == mce_start + 12)
+			monmsg->pos = mon_rec_start(monmsg);
+		goto out_copy;
+	}
+
+	/* read records */
+	if (monmsg->pos <= mon_rec_end(monmsg)) {
+		count = min(count, (size_t) mon_rec_end(monmsg) - monmsg->pos
+					    + 1);
+		ret = copy_to_user(data, (void *) (unsigned long) monmsg->pos,
+				   count);
+		if (ret)
+			return -EFAULT;
+		monmsg->pos += count;
+		if (monmsg->pos > mon_rec_end(monmsg))
+			mon_next_mca(monmsg);
+		goto out_copy;
+	}
+reply:
+	ret = mon_send_reply(monmsg, monpriv);
+	return ret;
+
+out_copy:
+	*ppos += count;
+	return count;
+}
+
+static unsigned int mon_poll(struct file *filp, struct poll_table_struct *p)
+{
+	struct mon_private *monpriv = filp->private_data;
+
+	poll_wait(filp, &mon_read_wait_queue, p);
+	if (unlikely(atomic_read(&monpriv->iucv_severed)))
+		return POLLERR;
+	if (atomic_read(&monpriv->read_ready))
+		return POLLIN | POLLRDNORM;
+	return 0;
+}
+
+static const struct file_operations mon_fops = {
+	.owner   = THIS_MODULE,
+	.open    = &mon_open,
+	.release = &mon_close,
+	.read    = &mon_read,
+	.poll    = &mon_poll,
+};
+
+static struct miscdevice mon_dev = {
+	.name       = "monreader",
+	.fops       = &mon_fops,
+	.minor      = MISC_DYNAMIC_MINOR,
+};
+
+
+/******************************************************************************
+ *				suspend / resume			      *
+ *****************************************************************************/
+static int monreader_freeze(struct device *dev)
+{
+	struct mon_private *monpriv = dev_get_drvdata(dev);
+	int rc;
+
+	if (!monpriv)
+		return 0;
+	if (monpriv->path) {
+		rc = iucv_path_sever(monpriv->path, user_data_sever);
+		if (rc)
+			pr_warning("Disconnecting the z/VM *MONITOR system "
+				   "service failed with rc=%i\n", rc);
+		iucv_path_free(monpriv->path);
+	}
+	atomic_set(&monpriv->iucv_severed, 0);
+	atomic_set(&monpriv->iucv_connected, 0);
+	atomic_set(&monpriv->read_ready, 0);
+	atomic_set(&monpriv->msglim_count, 0);
+	monpriv->write_index  = 0;
+	monpriv->read_index   = 0;
+	monpriv->path = NULL;
+	return 0;
+}
+
+static int monreader_thaw(struct device *dev)
+{
+	struct mon_private *monpriv = dev_get_drvdata(dev);
+	int rc;
+
+	if (!monpriv)
+		return 0;
+	rc = -ENOMEM;
+	monpriv->path = iucv_path_alloc(MON_MSGLIM, IUCV_IPRMDATA, GFP_KERNEL);
+	if (!monpriv->path)
+		goto out;
+	rc = iucv_path_connect(monpriv->path, &monreader_iucv_handler,
+			       MON_SERVICE, NULL, user_data_connect, monpriv);
+	if (rc) {
+		pr_err("Connecting to the z/VM *MONITOR system service "
+		       "failed with rc=%i\n", rc);
+		goto out_path;
+	}
+	wait_event(mon_conn_wait_queue,
+		   atomic_read(&monpriv->iucv_connected) ||
+		   atomic_read(&monpriv->iucv_severed));
+	if (atomic_read(&monpriv->iucv_severed))
+		goto out_path;
+	return 0;
+out_path:
+	rc = -EIO;
+	iucv_path_free(monpriv->path);
+	monpriv->path = NULL;
+out:
+	atomic_set(&monpriv->iucv_severed, 1);
+	return rc;
+}
+
+static int monreader_restore(struct device *dev)
+{
+	int rc;
+
+	segment_unload(mon_dcss_name);
+	rc = segment_load(mon_dcss_name, SEGMENT_SHARED,
+			  &mon_dcss_start, &mon_dcss_end);
+	if (rc < 0) {
+		segment_warning(rc, mon_dcss_name);
+		panic("fatal monreader resume error: no monitor dcss\n");
+	}
+	return monreader_thaw(dev);
+}
+
+static struct dev_pm_ops monreader_pm_ops = {
+	.freeze  = monreader_freeze,
+	.thaw	 = monreader_thaw,
+	.restore = monreader_restore,
+};
+
+static struct device_driver monreader_driver = {
+	.name = "monreader",
+	.bus  = &iucv_bus,
+	.pm   = &monreader_pm_ops,
+};
+
+
+/******************************************************************************
+ *                              module init/exit                              *
+ *****************************************************************************/
+static int __init mon_init(void)
+{
+	int rc;
+
+	if (!MACHINE_IS_VM) {
+		pr_err("The z/VM *MONITOR record device driver cannot be "
+		       "loaded without z/VM\n");
+		return -ENODEV;
+	}
+
+	/*
+	 * Register with IUCV and connect to *MONITOR service
+	 */
+	rc = iucv_register(&monreader_iucv_handler, 1);
+	if (rc) {
+		pr_err("The z/VM *MONITOR record device driver failed to "
+		       "register with IUCV\n");
+		return rc;
+	}
+
+	rc = driver_register(&monreader_driver);
+	if (rc)
+		goto out_iucv;
+	monreader_device = kzalloc(sizeof(struct device), GFP_KERNEL);
+	if (!monreader_device)
+		goto out_driver;
+	dev_set_name(monreader_device, "monreader-dev");
+	monreader_device->bus = &iucv_bus;
+	monreader_device->parent = iucv_root;
+	monreader_device->driver = &monreader_driver;
+	monreader_device->release = (void (*)(struct device *))kfree;
+	rc = device_register(monreader_device);
+	if (rc) {
+		put_device(monreader_device);
+		goto out_driver;
+	}
+
+	rc = segment_type(mon_dcss_name);
+	if (rc < 0) {
+		segment_warning(rc, mon_dcss_name);
+		goto out_device;
+	}
+	if (rc != SEG_TYPE_SC) {
+		pr_err("The specified *MONITOR DCSS %s does not have the "
+		       "required type SC\n", mon_dcss_name);
+		rc = -EINVAL;
+		goto out_device;
+	}
+
+	rc = segment_load(mon_dcss_name, SEGMENT_SHARED,
+			  &mon_dcss_start, &mon_dcss_end);
+	if (rc < 0) {
+		segment_warning(rc, mon_dcss_name);
+		rc = -EINVAL;
+		goto out_device;
+	}
+	dcss_mkname(mon_dcss_name, &user_data_connect[8]);
+
+	rc = misc_register(&mon_dev);
+	if (rc < 0 )
+		goto out;
+	return 0;
+
+out:
+	segment_unload(mon_dcss_name);
+out_device:
+	device_unregister(monreader_device);
+out_driver:
+	driver_unregister(&monreader_driver);
+out_iucv:
+	iucv_unregister(&monreader_iucv_handler, 1);
+	return rc;
+}
+
+static void __exit mon_exit(void)
+{
+	segment_unload(mon_dcss_name);
+	WARN_ON(misc_deregister(&mon_dev) != 0);
+	device_unregister(monreader_device);
+	driver_unregister(&monreader_driver);
+	iucv_unregister(&monreader_iucv_handler, 1);
+	return;
+}
+
+
+module_init(mon_init);
+module_exit(mon_exit);
+
+module_param_string(mondcss, mon_dcss_name, 9, 0444);
+MODULE_PARM_DESC(mondcss, "Name of DCSS segment to be used for *MONITOR "
+		 "service, max. 8 chars. Default is MONDCSS");
+
+MODULE_AUTHOR("Gerald Schaefer <geraldsc@de.ibm.com>");
+MODULE_DESCRIPTION("Character device driver for reading z/VM "
+		   "monitor service records.");
+MODULE_LICENSE("GPL");

kernel/drivers/s390/char/monwriter.c

@@ -0,0 +1,396 @@
+/*
+ * Character device driver for writing z/VM *MONITOR service records.
+ *
+ * Copyright IBM Corp. 2006, 2009
+ *
+ * Author(s): Melissa Howland <Melissa.Howland@us.ibm.com>
+ */
+
+#define KMSG_COMPONENT "monwriter"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/smp_lock.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/ctype.h>
+#include <linux/poll.h>
+#include <linux/mutex.h>
+#include <linux/platform_device.h>
+#include <asm/uaccess.h>
+#include <asm/ebcdic.h>
+#include <asm/io.h>
+#include <asm/appldata.h>
+#include <asm/monwriter.h>
+
+#define MONWRITE_MAX_DATALEN	4010
+
+static int mon_max_bufs = 255;
+static int mon_buf_count;
+
+struct mon_buf {
+	struct list_head list;
+	struct monwrite_hdr hdr;
+	int diag_done;
+	char *data;
+};
+
+static LIST_HEAD(mon_priv_list);
+
+struct mon_private {
+	struct list_head priv_list;
+	struct list_head list;
+	struct monwrite_hdr hdr;
+	size_t hdr_to_read;
+	size_t data_to_read;
+	struct mon_buf *current_buf;
+	struct mutex thread_mutex;
+};
+
+/*
+ * helper functions
+ */
+
+static int monwrite_diag(struct monwrite_hdr *myhdr, char *buffer, int fcn)
+{
+	struct appldata_product_id id;
+	int rc;
+
+	strcpy(id.prod_nr, "LNXAPPL");
+	id.prod_fn = myhdr->applid;
+	id.record_nr = myhdr->record_num;
+	id.version_nr = myhdr->version;
+	id.release_nr = myhdr->release;
+	id.mod_lvl = myhdr->mod_level;
+	rc = appldata_asm(&id, fcn, (void *) buffer, myhdr->datalen);
+	if (rc <= 0)
+		return rc;
+	pr_err("Writing monitor data failed with rc=%i\n", rc);
+	if (rc == 5)
+		return -EPERM;
+	return -EINVAL;
+}
+
+static struct mon_buf *monwrite_find_hdr(struct mon_private *monpriv,
+					 struct monwrite_hdr *monhdr)
+{
+	struct mon_buf *entry, *next;
+
+	list_for_each_entry_safe(entry, next, &monpriv->list, list)
+		if ((entry->hdr.mon_function == monhdr->mon_function ||
+		     monhdr->mon_function == MONWRITE_STOP_INTERVAL) &&
+		    entry->hdr.applid == monhdr->applid &&
+		    entry->hdr.record_num == monhdr->record_num &&
+		    entry->hdr.version == monhdr->version &&
+		    entry->hdr.release == monhdr->release &&
+		    entry->hdr.mod_level == monhdr->mod_level)
+			return entry;
+
+	return NULL;
+}
+
+static int monwrite_new_hdr(struct mon_private *monpriv)
+{
+	struct monwrite_hdr *monhdr = &monpriv->hdr;
+	struct mon_buf *monbuf;
+	int rc;
+
+	if (monhdr->datalen > MONWRITE_MAX_DATALEN ||
+	    monhdr->mon_function > MONWRITE_START_CONFIG ||
+	    monhdr->hdrlen != sizeof(struct monwrite_hdr))
+		return -EINVAL;
+	monbuf = NULL;
+	if (monhdr->mon_function != MONWRITE_GEN_EVENT)
+		monbuf = monwrite_find_hdr(monpriv, monhdr);
+	if (monbuf) {
+		if (monhdr->mon_function == MONWRITE_STOP_INTERVAL) {
+			monhdr->datalen = monbuf->hdr.datalen;
+			rc = monwrite_diag(monhdr, monbuf->data,
+					   APPLDATA_STOP_REC);
+			list_del(&monbuf->list);
+			mon_buf_count--;
+			kfree(monbuf->data);
+			kfree(monbuf);
+			monbuf = NULL;
+		}
+	} else if (monhdr->mon_function != MONWRITE_STOP_INTERVAL) {
+		if (mon_buf_count >= mon_max_bufs)
+			return -ENOSPC;
+		monbuf = kzalloc(sizeof(struct mon_buf), GFP_KERNEL);
+		if (!monbuf)
+			return -ENOMEM;
+		monbuf->data = kzalloc(monhdr->datalen,
+				       GFP_KERNEL | GFP_DMA);
+		if (!monbuf->data) {
+			kfree(monbuf);
+			return -ENOMEM;
+		}
+		monbuf->hdr = *monhdr;
+		list_add_tail(&monbuf->list, &monpriv->list);
+		if (monhdr->mon_function != MONWRITE_GEN_EVENT)
+			mon_buf_count++;
+	}
+	monpriv->current_buf = monbuf;
+	return 0;
+}
+
+static int monwrite_new_data(struct mon_private *monpriv)
+{
+	struct monwrite_hdr *monhdr = &monpriv->hdr;
+	struct mon_buf *monbuf = monpriv->current_buf;
+	int rc = 0;
+
+	switch (monhdr->mon_function) {
+	case MONWRITE_START_INTERVAL:
+		if (!monbuf->diag_done) {
+			rc = monwrite_diag(monhdr, monbuf->data,
+					   APPLDATA_START_INTERVAL_REC);
+			monbuf->diag_done = 1;
+		}
+		break;
+	case MONWRITE_START_CONFIG:
+		if (!monbuf->diag_done) {
+			rc = monwrite_diag(monhdr, monbuf->data,
+					   APPLDATA_START_CONFIG_REC);
+			monbuf->diag_done = 1;
+		}
+		break;
+	case MONWRITE_GEN_EVENT:
+		rc = monwrite_diag(monhdr, monbuf->data,
+				   APPLDATA_GEN_EVENT_REC);
+		list_del(&monpriv->current_buf->list);
+		kfree(monpriv->current_buf->data);
+		kfree(monpriv->current_buf);
+		monpriv->current_buf = NULL;
+		break;
+	default:
+		/* monhdr->mon_function is checked in monwrite_new_hdr */
+		BUG();
+	}
+	return rc;
+}
+
+/*
+ * file operations
+ */
+
+static int monwrite_open(struct inode *inode, struct file *filp)
+{
+	struct mon_private *monpriv;
+
+	monpriv = kzalloc(sizeof(struct mon_private), GFP_KERNEL);
+	if (!monpriv)
+		return -ENOMEM;
+	lock_kernel();
+	INIT_LIST_HEAD(&monpriv->list);
+	monpriv->hdr_to_read = sizeof(monpriv->hdr);
+	mutex_init(&monpriv->thread_mutex);
+	filp->private_data = monpriv;
+	list_add_tail(&monpriv->priv_list, &mon_priv_list);
+	unlock_kernel();
+	return nonseekable_open(inode, filp);
+}
+
+static int monwrite_close(struct inode *inode, struct file *filp)
+{
+	struct mon_private *monpriv = filp->private_data;
+	struct mon_buf *entry, *next;
+
+	list_for_each_entry_safe(entry, next, &monpriv->list, list) {
+		if (entry->hdr.mon_function != MONWRITE_GEN_EVENT)
+			monwrite_diag(&entry->hdr, entry->data,
+				      APPLDATA_STOP_REC);
+		mon_buf_count--;
+		list_del(&entry->list);
+		kfree(entry->data);
+		kfree(entry);
+	}
+	list_del(&monpriv->priv_list);
+	kfree(monpriv);
+	return 0;
+}
+
+static ssize_t monwrite_write(struct file *filp, const char __user *data,
+			      size_t count, loff_t *ppos)
+{
+	struct mon_private *monpriv = filp->private_data;
+	size_t len, written;
+	void *to;
+	int rc;
+
+	mutex_lock(&monpriv->thread_mutex);
+	for (written = 0; written < count; ) {
+		if (monpriv->hdr_to_read) {
+			len = min(count - written, monpriv->hdr_to_read);
+			to = (char *) &monpriv->hdr +
+				sizeof(monpriv->hdr) - monpriv->hdr_to_read;
+			if (copy_from_user(to, data + written, len)) {
+				rc = -EFAULT;
+				goto out_error;
+			}
+			monpriv->hdr_to_read -= len;
+			written += len;
+			if (monpriv->hdr_to_read > 0)
+				continue;
+			rc = monwrite_new_hdr(monpriv);
+			if (rc)
+				goto out_error;
+			monpriv->data_to_read = monpriv->current_buf ?
+				monpriv->current_buf->hdr.datalen : 0;
+		}
+
+		if (monpriv->data_to_read) {
+			len = min(count - written, monpriv->data_to_read);
+			to = monpriv->current_buf->data +
+				monpriv->hdr.datalen - monpriv->data_to_read;
+			if (copy_from_user(to, data + written, len)) {
+				rc = -EFAULT;
+				goto out_error;
+			}
+			monpriv->data_to_read -= len;
+			written += len;
+			if (monpriv->data_to_read > 0)
+				continue;
+			rc = monwrite_new_data(monpriv);
+			if (rc)
+				goto out_error;
+		}
+		monpriv->hdr_to_read = sizeof(monpriv->hdr);
+	}
+	mutex_unlock(&monpriv->thread_mutex);
+	return written;
+
+out_error:
+	monpriv->data_to_read = 0;
+	monpriv->hdr_to_read = sizeof(struct monwrite_hdr);
+	mutex_unlock(&monpriv->thread_mutex);
+	return rc;
+}
+
+static const struct file_operations monwrite_fops = {
+	.owner	 = THIS_MODULE,
+	.open	 = &monwrite_open,
+	.release = &monwrite_close,
+	.write	 = &monwrite_write,
+};
+
+static struct miscdevice mon_dev = {
+	.name	= "monwriter",
+	.fops	= &monwrite_fops,
+	.minor	= MISC_DYNAMIC_MINOR,
+};
+
+/*
+ * suspend/resume
+ */
+
+static int monwriter_freeze(struct device *dev)
+{
+	struct mon_private *monpriv;
+	struct mon_buf *monbuf;
+
+	list_for_each_entry(monpriv, &mon_priv_list, priv_list) {
+		list_for_each_entry(monbuf, &monpriv->list, list) {
+			if (monbuf->hdr.mon_function != MONWRITE_GEN_EVENT)
+				monwrite_diag(&monbuf->hdr, monbuf->data,
+					      APPLDATA_STOP_REC);
+		}
+	}
+	return 0;
+}
+
+static int monwriter_restore(struct device *dev)
+{
+	struct mon_private *monpriv;
+	struct mon_buf *monbuf;
+
+	list_for_each_entry(monpriv, &mon_priv_list, priv_list) {
+		list_for_each_entry(monbuf, &monpriv->list, list) {
+			if (monbuf->hdr.mon_function == MONWRITE_START_INTERVAL)
+				monwrite_diag(&monbuf->hdr, monbuf->data,
+					      APPLDATA_START_INTERVAL_REC);
+			if (monbuf->hdr.mon_function == MONWRITE_START_CONFIG)
+				monwrite_diag(&monbuf->hdr, monbuf->data,
+					      APPLDATA_START_CONFIG_REC);
+		}
+	}
+	return 0;
+}
+
+static int monwriter_thaw(struct device *dev)
+{
+	return monwriter_restore(dev);
+}
+
+static struct dev_pm_ops monwriter_pm_ops = {
+	.freeze		= monwriter_freeze,
+	.thaw		= monwriter_thaw,
+	.restore	= monwriter_restore,
+};
+
+static struct platform_driver monwriter_pdrv = {
+	.driver = {
+		.name	= "monwriter",
+		.owner	= THIS_MODULE,
+		.pm	= &monwriter_pm_ops,
+	},
+};
+
+static struct platform_device *monwriter_pdev;
+
+/*
+ * module init/exit
+ */
+
+static int __init mon_init(void)
+{
+	int rc;
+
+	if (!MACHINE_IS_VM)
+		return -ENODEV;
+
+	rc = platform_driver_register(&monwriter_pdrv);
+	if (rc)
+		return rc;
+
+	monwriter_pdev = platform_device_register_simple("monwriter", -1, NULL,
+							0);
+	if (IS_ERR(monwriter_pdev)) {
+		rc = PTR_ERR(monwriter_pdev);
+		goto out_driver;
+	}
+
+	rc = misc_register(&mon_dev);
+	if (rc)
+		goto out_device;
+	return 0;
+
+out_device:
+	platform_device_unregister(monwriter_pdev);
+out_driver:
+	platform_driver_unregister(&monwriter_pdrv);
+	return rc;
+}
+
+static void __exit mon_exit(void)
+{
+	WARN_ON(misc_deregister(&mon_dev) != 0);
+	platform_device_unregister(monwriter_pdev);
+	platform_driver_unregister(&monwriter_pdrv);
+}
+
+module_init(mon_init);
+module_exit(mon_exit);
+
+module_param_named(max_bufs, mon_max_bufs, int, 0644);
+MODULE_PARM_DESC(max_bufs, "Maximum number of sample monitor data buffers "
+		 "that can be active at one time");
+
+MODULE_AUTHOR("Melissa Howland <Melissa.Howland@us.ibm.com>");
+MODULE_DESCRIPTION("Character device driver for writing z/VM "
+		   "APPLDATA monitor records.");
+MODULE_LICENSE("GPL");

kernel/drivers/s390/char/raw3270.h

@@ -0,0 +1,279 @@
+/*
+ * IBM/3270 Driver
+ *
+ * Author(s):
+ *   Original 3270 Code for 2.4 written by Richard Hitt (UTS Global)
+ *   Rewritten for 2.5 by Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *     Copyright IBM Corp. 2003, 2009
+ */
+
+#include <asm/idals.h>
+#include <asm/ioctl.h>
+
+/* ioctls for fullscreen 3270 */
+#define TUBICMD		_IO('3', 3)	/* set ccw command for fs reads. */
+#define TUBOCMD		_IO('3', 4)	/* set ccw command for fs writes. */
+#define TUBGETI		_IO('3', 7)	/* get ccw command for fs reads. */
+#define TUBGETO		_IO('3', 8)	/* get ccw command for fs writes. */
+#define TUBSETMOD	_IO('3',12)	/* FIXME: what does it do ?*/
+#define TUBGETMOD	_IO('3',13)	/* FIXME: what does it do ?*/
+
+/* Local Channel Commands */
+#define TC_WRITE	0x01		/* Write */
+#define TC_RDBUF	0x02		/* Read Buffer */
+#define TC_EWRITE	0x05		/* Erase write */
+#define TC_READMOD	0x06		/* Read modified */
+#define TC_EWRITEA	0x0d		/* Erase write alternate */
+#define TC_WRITESF	0x11		/* Write structured field */
+
+/* Buffer Control Orders */
+#define TO_SF		0x1d		/* Start field */
+#define TO_SBA		0x11		/* Set buffer address */
+#define TO_IC		0x13		/* Insert cursor */
+#define TO_PT		0x05		/* Program tab */
+#define TO_RA		0x3c		/* Repeat to address */
+#define TO_SFE		0x29		/* Start field extended */
+#define TO_EUA		0x12		/* Erase unprotected to address */
+#define TO_MF		0x2c		/* Modify field */
+#define TO_SA		0x28		/* Set attribute */
+
+/* Field Attribute Bytes */
+#define TF_INPUT	0x40		/* Visible input */
+#define TF_INPUTN	0x4c		/* Invisible input */
+#define TF_INMDT	0xc1		/* Visible, Set-MDT */
+#define TF_LOG		0x60
+
+/* Character Attribute Bytes */
+#define TAT_RESET	0x00
+#define TAT_FIELD	0xc0
+#define TAT_EXTHI	0x41
+#define TAT_COLOR	0x42
+#define TAT_CHARS	0x43
+#define TAT_TRANS	0x46
+
+/* Extended-Highlighting Bytes */
+#define TAX_RESET	0x00
+#define TAX_BLINK	0xf1
+#define TAX_REVER	0xf2
+#define TAX_UNDER	0xf4
+
+/* Reset value */
+#define TAR_RESET	0x00
+
+/* Color values */
+#define TAC_RESET	0x00
+#define TAC_BLUE	0xf1
+#define TAC_RED		0xf2
+#define TAC_PINK	0xf3
+#define TAC_GREEN	0xf4
+#define TAC_TURQ	0xf5
+#define TAC_YELLOW	0xf6
+#define TAC_WHITE	0xf7
+#define TAC_DEFAULT	0x00
+
+/* Write Control Characters */
+#define TW_NONE		0x40		/* No particular action */
+#define TW_KR		0xc2		/* Keyboard restore */
+#define TW_PLUSALARM	0x04		/* Add this bit for alarm */
+
+#define RAW3270_FIRSTMINOR	1	/* First minor number */
+#define RAW3270_MAXDEVS		255	/* Max number of 3270 devices */
+
+/* For TUBGETMOD and TUBSETMOD. Should include. */
+struct raw3270_iocb {
+	short model;
+	short line_cnt;
+	short col_cnt;
+	short pf_cnt;
+	short re_cnt;
+	short map;
+};
+
+struct raw3270;
+struct raw3270_view;
+
+/* 3270 CCW request */
+struct raw3270_request {
+	struct list_head list;		/* list head for request queueing. */
+	struct raw3270_view *view;	/* view of this request */
+	struct ccw1 ccw;		/* single ccw. */
+	void *buffer;			/* output buffer. */
+	size_t size;			/* size of output buffer. */
+	int rescnt;			/* residual count from devstat. */
+	int rc;				/* return code for this request. */
+
+	/* Callback for delivering final status. */
+	void (*callback)(struct raw3270_request *, void *);
+	void *callback_data;
+};
+
+struct raw3270_request *raw3270_request_alloc(size_t size);
+struct raw3270_request *raw3270_request_alloc_bootmem(size_t size);
+void raw3270_request_free(struct raw3270_request *);
+void raw3270_request_reset(struct raw3270_request *);
+void raw3270_request_set_cmd(struct raw3270_request *, u8 cmd);
+int  raw3270_request_add_data(struct raw3270_request *, void *, size_t);
+void raw3270_request_set_data(struct raw3270_request *, void *, size_t);
+void raw3270_request_set_idal(struct raw3270_request *, struct idal_buffer *);
+
+static inline int
+raw3270_request_final(struct raw3270_request *rq)
+{
+	return list_empty(&rq->list);
+}
+
+void raw3270_buffer_address(struct raw3270 *, char *, unsigned short);
+
+/* Return value of *intv (see raw3270_fn below) can be one of the following: */
+#define RAW3270_IO_DONE		0	/* request finished */
+#define RAW3270_IO_BUSY		1	/* request still active */
+#define RAW3270_IO_RETRY	2	/* retry current request */
+#define RAW3270_IO_STOP		3	/* kill current request */
+
+/*
+ * Functions of a 3270 view.
+ */
+struct raw3270_fn {
+	int  (*activate)(struct raw3270_view *);
+	void (*deactivate)(struct raw3270_view *);
+	int  (*intv)(struct raw3270_view *,
+		     struct raw3270_request *, struct irb *);
+	void (*release)(struct raw3270_view *);
+	void (*free)(struct raw3270_view *);
+};
+
+/*
+ * View structure chaining. The raw3270_view structure is meant to
+ * be embedded at the start of the real view data structure, e.g.:
+ *   struct example {
+ *     struct raw3270_view view;
+ *     ...
+ *   };
+ */
+struct raw3270_view {
+	struct list_head list;
+	spinlock_t lock;
+	atomic_t ref_count;
+	struct raw3270 *dev;
+	struct raw3270_fn *fn;
+	unsigned int model;
+	unsigned int rows, cols;	/* # of rows & colums of the view */
+	unsigned char *ascebc;		/* ascii -> ebcdic table */
+};
+
+int raw3270_add_view(struct raw3270_view *, struct raw3270_fn *, int);
+int raw3270_activate_view(struct raw3270_view *);
+void raw3270_del_view(struct raw3270_view *);
+void raw3270_deactivate_view(struct raw3270_view *);
+struct raw3270_view *raw3270_find_view(struct raw3270_fn *, int);
+int raw3270_start(struct raw3270_view *, struct raw3270_request *);
+int raw3270_start_locked(struct raw3270_view *, struct raw3270_request *);
+int raw3270_start_irq(struct raw3270_view *, struct raw3270_request *);
+int raw3270_reset(struct raw3270_view *);
+struct raw3270_view *raw3270_view(struct raw3270_view *);
+
+/* Reference count inliner for view structures. */
+static inline void
+raw3270_get_view(struct raw3270_view *view)
+{
+	atomic_inc(&view->ref_count);
+}
+
+extern wait_queue_head_t raw3270_wait_queue;
+
+static inline void
+raw3270_put_view(struct raw3270_view *view)
+{
+	if (atomic_dec_return(&view->ref_count) == 0)
+		wake_up(&raw3270_wait_queue);
+}
+
+struct raw3270 *raw3270_setup_console(struct ccw_device *cdev);
+void raw3270_wait_cons_dev(struct raw3270 *);
+
+/* Notifier for device addition/removal */
+int raw3270_register_notifier(void (*notifier)(int, int));
+void raw3270_unregister_notifier(void (*notifier)(int, int));
+void raw3270_pm_unfreeze(struct raw3270_view *);
+
+/*
+ * Little memory allocator for string objects. 
+ */
+struct string
+{
+	struct list_head list;
+	struct list_head update;
+	unsigned long size;
+	unsigned long len;
+	char string[0];
+} __attribute__ ((aligned(8)));
+
+static inline struct string *
+alloc_string(struct list_head *free_list, unsigned long len)
+{
+	struct string *cs, *tmp;
+	unsigned long size;
+
+	size = (len + 7L) & -8L;
+	list_for_each_entry(cs, free_list, list) {
+		if (cs->size < size)
+			continue;
+		if (cs->size > size + sizeof(struct string)) {
+			char *endaddr = (char *) (cs + 1) + cs->size;
+			tmp = (struct string *) (endaddr - size) - 1;
+			tmp->size = size;
+			cs->size -= size + sizeof(struct string);
+			cs = tmp;
+		} else
+			list_del(&cs->list);
+		cs->len = len;
+		INIT_LIST_HEAD(&cs->list);
+		INIT_LIST_HEAD(&cs->update);
+		return cs;
+	}
+	return NULL;
+}
+
+static inline unsigned long
+free_string(struct list_head *free_list, struct string *cs)
+{
+	struct string *tmp;
+	struct list_head *p, *left;
+
+	/* Find out the left neighbour in free memory list. */
+	left = free_list;
+	list_for_each(p, free_list) {
+		if (list_entry(p, struct string, list) > cs)
+			break;
+		left = p;
+	}
+	/* Try to merge with right neighbour = next element from left. */
+	if (left->next != free_list) {
+		tmp = list_entry(left->next, struct string, list);
+		if ((char *) (cs + 1) + cs->size == (char *) tmp) {
+			list_del(&tmp->list);
+			cs->size += tmp->size + sizeof(struct string);
+		}
+	}
+	/* Try to merge with left neighbour. */
+	if (left != free_list) {
+		tmp = list_entry(left, struct string, list);
+		if ((char *) (tmp + 1) + tmp->size == (char *) cs) {
+			tmp->size += cs->size + sizeof(struct string);
+			return tmp->size;
+		}
+	}
+	__list_add(&cs->list, left, left->next);
+	return cs->size;
+}
+
+static inline void
+add_string_memory(struct list_head *free_list, void *mem, unsigned long size)
+{
+	struct string *cs;
+
+	cs = (struct string *) mem;
+	cs->size = size - sizeof(struct string);
+	free_string(free_list, cs);
+}
+

kernel/drivers/s390/char/sclp.h

@@ -0,0 +1,189 @@
+/*
+ * Copyright IBM Corp. 1999, 2009
+ *
+ * Author(s): Martin Peschke <mpeschke@de.ibm.com>
+ *	      Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#ifndef __SCLP_H__
+#define __SCLP_H__
+
+#include <linux/types.h>
+#include <linux/list.h>
+#include <asm/sclp.h>
+#include <asm/ebcdic.h>
+
+/* maximum number of pages concerning our own memory management */
+#define MAX_KMEM_PAGES (sizeof(unsigned long) << 3)
+#define MAX_CONSOLE_PAGES	6
+
+#define EVTYP_OPCMD		0x01
+#define EVTYP_MSG		0x02
+#define EVTYP_STATECHANGE	0x08
+#define EVTYP_PMSGCMD		0x09
+#define EVTYP_CNTLPROGOPCMD	0x20
+#define EVTYP_CNTLPROGIDENT	0x0B
+#define EVTYP_SIGQUIESCE	0x1D
+#define EVTYP_VT220MSG		0x1A
+#define EVTYP_CONFMGMDATA	0x04
+#define EVTYP_SDIAS		0x1C
+#define EVTYP_ASYNC		0x0A
+
+#define EVTYP_OPCMD_MASK	0x80000000
+#define EVTYP_MSG_MASK		0x40000000
+#define EVTYP_STATECHANGE_MASK	0x01000000
+#define EVTYP_PMSGCMD_MASK	0x00800000
+#define EVTYP_CTLPROGOPCMD_MASK	0x00000001
+#define EVTYP_CTLPROGIDENT_MASK	0x00200000
+#define EVTYP_SIGQUIESCE_MASK	0x00000008
+#define EVTYP_VT220MSG_MASK	0x00000040
+#define EVTYP_CONFMGMDATA_MASK	0x10000000
+#define EVTYP_SDIAS_MASK	0x00000010
+#define EVTYP_ASYNC_MASK	0x00400000
+
+#define GNRLMSGFLGS_DOM		0x8000
+#define GNRLMSGFLGS_SNDALRM	0x4000
+#define GNRLMSGFLGS_HOLDMSG	0x2000
+
+#define LNTPFLGS_CNTLTEXT	0x8000
+#define LNTPFLGS_LABELTEXT	0x4000
+#define LNTPFLGS_DATATEXT	0x2000
+#define LNTPFLGS_ENDTEXT	0x1000
+#define LNTPFLGS_PROMPTTEXT	0x0800
+
+typedef unsigned int sclp_cmdw_t;
+
+#define SCLP_CMDW_READ_EVENT_DATA	0x00770005
+#define SCLP_CMDW_WRITE_EVENT_DATA	0x00760005
+#define SCLP_CMDW_WRITE_EVENT_MASK	0x00780005
+
+#define GDS_ID_MDSMU		0x1310
+#define GDS_ID_MDSROUTEINFO	0x1311
+#define GDS_ID_AGUNWRKCORR	0x1549
+#define GDS_ID_SNACONDREPORT	0x1532
+#define GDS_ID_CPMSU		0x1212
+#define GDS_ID_ROUTTARGINSTR	0x154D
+#define GDS_ID_OPREQ		0x8070
+#define GDS_ID_TEXTCMD		0x1320
+
+#define GDS_KEY_SELFDEFTEXTMSG	0x31
+
+enum sclp_pm_event {
+	SCLP_PM_EVENT_FREEZE,
+	SCLP_PM_EVENT_THAW,
+	SCLP_PM_EVENT_RESTORE,
+};
+
+#define SCLP_PANIC_PRIO		1
+#define SCLP_PANIC_PRIO_CLIENT	0
+
+typedef u32 sccb_mask_t;	/* ATTENTION: assumes 32bit mask !!! */
+
+struct sccb_header {
+	u16	length;
+	u8	function_code;
+	u8	control_mask[3];
+	u16	response_code;
+} __attribute__((packed));
+
+extern u64 sclp_facilities;
+#define SCLP_HAS_CHP_INFO	(sclp_facilities & 0x8000000000000000ULL)
+#define SCLP_HAS_CHP_RECONFIG	(sclp_facilities & 0x2000000000000000ULL)
+#define SCLP_HAS_CPU_INFO	(sclp_facilities & 0x0800000000000000ULL)
+#define SCLP_HAS_CPU_RECONFIG	(sclp_facilities & 0x0400000000000000ULL)
+
+
+struct gds_subvector {
+	u8	length;
+	u8	key;
+} __attribute__((packed));
+
+struct gds_vector {
+	u16	length;
+	u16	gds_id;
+} __attribute__((packed));
+
+struct evbuf_header {
+	u16	length;
+	u8	type;
+	u8	flags;
+	u16	_reserved;
+} __attribute__((packed));
+
+struct sclp_req {
+	struct list_head list;		/* list_head for request queueing. */
+	sclp_cmdw_t command;		/* sclp command to execute */
+	void	*sccb;			/* pointer to the sccb to execute */
+	char	status;			/* status of this request */
+	int     start_count;		/* number of SVCs done for this req */
+	/* Callback that is called after reaching final status. */
+	void (*callback)(struct sclp_req *, void *data);
+	void *callback_data;
+};
+
+#define SCLP_REQ_FILLED	  0x00	/* request is ready to be processed */
+#define SCLP_REQ_QUEUED	  0x01	/* request is queued to be processed */
+#define SCLP_REQ_RUNNING  0x02	/* request is currently running */
+#define SCLP_REQ_DONE	  0x03	/* request is completed successfully */
+#define SCLP_REQ_FAILED	  0x05	/* request is finally failed */
+
+/* function pointers that a high level driver has to use for registration */
+/* of some routines it wants to be called from the low level driver */
+struct sclp_register {
+	struct list_head list;
+	/* User wants to receive: */
+	sccb_mask_t receive_mask;
+	/* User wants to send: */
+	sccb_mask_t send_mask;
+	/* H/W can receive: */
+	sccb_mask_t sclp_receive_mask;
+	/* H/W can send: */
+	sccb_mask_t sclp_send_mask;
+	/* called if event type availability changes */
+	void (*state_change_fn)(struct sclp_register *);
+	/* called for events in cp_receive_mask/sclp_receive_mask */
+	void (*receiver_fn)(struct evbuf_header *);
+	/* called for power management events */
+	void (*pm_event_fn)(struct sclp_register *, enum sclp_pm_event);
+	/* pm event posted flag */
+	int pm_event_posted;
+};
+
+/* externals from sclp.c */
+int sclp_add_request(struct sclp_req *req);
+void sclp_sync_wait(void);
+int sclp_register(struct sclp_register *reg);
+void sclp_unregister(struct sclp_register *reg);
+int sclp_remove_processed(struct sccb_header *sccb);
+int sclp_deactivate(void);
+int sclp_reactivate(void);
+int sclp_service_call(sclp_cmdw_t command, void *sccb);
+
+int sclp_sdias_init(void);
+void sclp_sdias_exit(void);
+
+/* useful inlines */
+
+/* VM uses EBCDIC 037, LPAR+native(SE+HMC) use EBCDIC 500 */
+/* translate single character from ASCII to EBCDIC */
+static inline unsigned char
+sclp_ascebc(unsigned char ch)
+{
+	return (MACHINE_IS_VM) ? _ascebc[ch] : _ascebc_500[ch];
+}
+
+/* translate string from EBCDIC to ASCII */
+static inline void
+sclp_ebcasc_str(unsigned char *str, int nr)
+{
+	(MACHINE_IS_VM) ? EBCASC(str, nr) : EBCASC_500(str, nr);
+}
+
+/* translate string from ASCII to EBCDIC */
+static inline void
+sclp_ascebc_str(unsigned char *str, int nr)
+{
+	(MACHINE_IS_VM) ? ASCEBC(str, nr) : ASCEBC_500(str, nr);
+}
+
+#endif	 /* __SCLP_H__ */

kernel/drivers/s390/char/sclp_async.c

@@ -0,0 +1,213 @@
+/*
+ * Enable Asynchronous Notification via SCLP.
+ *
+ * Copyright IBM Corp. 2009
+ * Author(s): Hans-Joachim Picht <hans@linux.vnet.ibm.com>
+ *
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/stat.h>
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/kmod.h>
+#include <linux/err.h>
+#include <linux/errno.h>
+#include <linux/proc_fs.h>
+#include <linux/sysctl.h>
+#include <linux/utsname.h>
+#include "sclp.h"
+
+static int callhome_enabled;
+static struct sclp_req *request;
+static struct sclp_async_sccb *sccb;
+static int sclp_async_send_wait(char *message);
+static struct ctl_table_header *callhome_sysctl_header;
+static DEFINE_SPINLOCK(sclp_async_lock);
+#define SCLP_NORMAL_WRITE	0x00
+
+struct async_evbuf {
+	struct evbuf_header header;
+	u64 reserved;
+	u8 rflags;
+	u8 empty;
+	u8 rtype;
+	u8 otype;
+	char comp_id[12];
+	char data[3000]; /* there is still some space left */
+} __attribute__((packed));
+
+struct sclp_async_sccb {
+	struct sccb_header header;
+	struct async_evbuf evbuf;
+} __attribute__((packed));
+
+static struct sclp_register sclp_async_register = {
+	.send_mask = EVTYP_ASYNC_MASK,
+};
+
+static int call_home_on_panic(struct notifier_block *self,
+			      unsigned long event, void *data)
+{
+	strncat(data, init_utsname()->nodename,
+		sizeof(init_utsname()->nodename));
+	sclp_async_send_wait(data);
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block call_home_panic_nb = {
+	.notifier_call = call_home_on_panic,
+	.priority = INT_MAX,
+};
+
+static int proc_handler_callhome(struct ctl_table *ctl, int write,
+				 void __user *buffer, size_t *count,
+				 loff_t *ppos)
+{
+	unsigned long val;
+	int len, rc;
+	char buf[3];
+
+	if (!*count || (*ppos && !write)) {
+		*count = 0;
+		return 0;
+	}
+	if (!write) {
+		len = snprintf(buf, sizeof(buf), "%d\n", callhome_enabled);
+		rc = copy_to_user(buffer, buf, sizeof(buf));
+		if (rc != 0)
+			return -EFAULT;
+	} else {
+		len = *count;
+		rc = copy_from_user(buf, buffer, sizeof(buf));
+		if (rc != 0)
+			return -EFAULT;
+		if (strict_strtoul(buf, 0, &val) != 0)
+			return -EINVAL;
+		if (val != 0 && val != 1)
+			return -EINVAL;
+		callhome_enabled = val;
+	}
+	*count = len;
+	*ppos += len;
+	return 0;
+}
+
+static struct ctl_table callhome_table[] = {
+	{
+		.procname	= "callhome",
+		.mode		= 0644,
+		.proc_handler	= proc_handler_callhome,
+	},
+	{ .ctl_name = 0 }
+};
+
+static struct ctl_table kern_dir_table[] = {
+	{
+		.ctl_name	= CTL_KERN,
+		.procname	= "kernel",
+		.maxlen		= 0,
+		.mode		= 0555,
+		.child		= callhome_table,
+	},
+	{ .ctl_name = 0 }
+};
+
+/*
+ * Function used to transfer asynchronous notification
+ * records which waits for send completion
+ */
+static int sclp_async_send_wait(char *message)
+{
+	struct async_evbuf *evb;
+	int rc;
+	unsigned long flags;
+
+	if (!callhome_enabled)
+		return 0;
+	sccb->evbuf.header.type = EVTYP_ASYNC;
+	sccb->evbuf.rtype = 0xA5;
+	sccb->evbuf.otype = 0x00;
+	evb = &sccb->evbuf;
+	request->command = SCLP_CMDW_WRITE_EVENT_DATA;
+	request->sccb = sccb;
+	request->status = SCLP_REQ_FILLED;
+	strncpy(sccb->evbuf.data, message, sizeof(sccb->evbuf.data));
+	/*
+	 * Retain Queue
+	 * e.g. 5639CC140 500 Red Hat RHEL5 Linux for zSeries (RHEL AS)
+	 */
+	strncpy(sccb->evbuf.comp_id, "000000000", sizeof(sccb->evbuf.comp_id));
+	sccb->evbuf.header.length = sizeof(sccb->evbuf);
+	sccb->header.length = sizeof(sccb->evbuf) + sizeof(sccb->header);
+	sccb->header.function_code = SCLP_NORMAL_WRITE;
+	rc = sclp_add_request(request);
+	if (rc)
+		return rc;
+	spin_lock_irqsave(&sclp_async_lock, flags);
+	while (request->status != SCLP_REQ_DONE &&
+		request->status != SCLP_REQ_FAILED) {
+		 sclp_sync_wait();
+	}
+	spin_unlock_irqrestore(&sclp_async_lock, flags);
+	if (request->status != SCLP_REQ_DONE)
+		return -EIO;
+	rc = ((struct sclp_async_sccb *)
+	       request->sccb)->header.response_code;
+	if (rc != 0x0020)
+		return -EIO;
+	if (evb->header.flags != 0x80)
+		return -EIO;
+	return rc;
+}
+
+static int __init sclp_async_init(void)
+{
+	int rc;
+
+	rc = sclp_register(&sclp_async_register);
+	if (rc)
+		return rc;
+	rc = -EOPNOTSUPP;
+	if (!(sclp_async_register.sclp_receive_mask & EVTYP_ASYNC_MASK))
+		goto out_sclp;
+	rc = -ENOMEM;
+	callhome_sysctl_header = register_sysctl_table(kern_dir_table);
+	if (!callhome_sysctl_header)
+		goto out_sclp;
+	request = kzalloc(sizeof(struct sclp_req), GFP_KERNEL);
+	sccb = (struct sclp_async_sccb *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!request || !sccb)
+		goto out_mem;
+	rc = atomic_notifier_chain_register(&panic_notifier_list,
+					    &call_home_panic_nb);
+	if (!rc)
+		goto out;
+out_mem:
+	kfree(request);
+	free_page((unsigned long) sccb);
+	unregister_sysctl_table(callhome_sysctl_header);
+out_sclp:
+	sclp_unregister(&sclp_async_register);
+out:
+	return rc;
+}
+module_init(sclp_async_init);
+
+static void __exit sclp_async_exit(void)
+{
+	atomic_notifier_chain_unregister(&panic_notifier_list,
+					 &call_home_panic_nb);
+	unregister_sysctl_table(callhome_sysctl_header);
+	sclp_unregister(&sclp_async_register);
+	free_page((unsigned long) sccb);
+	kfree(request);
+}
+module_exit(sclp_async_exit);
+
+MODULE_AUTHOR("Copyright IBM Corp. 2009");
+MODULE_AUTHOR("Hans-Joachim Picht <hans@linux.vnet.ibm.com>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("SCLP Asynchronous Notification Records");

kernel/drivers/s390/char/sclp_cmd.c

@@ -0,0 +1,751 @@
+/*
+ * Copyright IBM Corp. 2007, 2009
+ *
+ * Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>,
+ *	      Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#define KMSG_COMPONENT "sclp_cmd"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/completion.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/mm.h>
+#include <linux/mmzone.h>
+#include <linux/memory.h>
+#include <linux/platform_device.h>
+#include <asm/chpid.h>
+#include <asm/sclp.h>
+#include <asm/setup.h>
+
+#include "sclp.h"
+
+#define SCLP_CMDW_READ_SCP_INFO		0x00020001
+#define SCLP_CMDW_READ_SCP_INFO_FORCED	0x00120001
+
+struct read_info_sccb {
+	struct	sccb_header header;	/* 0-7 */
+	u16	rnmax;			/* 8-9 */
+	u8	rnsize;			/* 10 */
+	u8	_reserved0[24 - 11];	/* 11-15 */
+	u8	loadparm[8];		/* 24-31 */
+	u8	_reserved1[48 - 32];	/* 32-47 */
+	u64	facilities;		/* 48-55 */
+	u8	_reserved2[84 - 56];	/* 56-83 */
+	u8	fac84;			/* 84 */
+	u8	_reserved3[91 - 85];	/* 85-90 */
+	u8	flags;			/* 91 */
+	u8	_reserved4[100 - 92];	/* 92-99 */
+	u32	rnsize2;		/* 100-103 */
+	u64	rnmax2;			/* 104-111 */
+	u8	_reserved5[4096 - 112];	/* 112-4095 */
+} __attribute__((packed, aligned(PAGE_SIZE)));
+
+static struct read_info_sccb __initdata early_read_info_sccb;
+static int __initdata early_read_info_sccb_valid;
+
+u64 sclp_facilities;
+static u8 sclp_fac84;
+static unsigned long long rzm;
+static unsigned long long rnmax;
+
+static int __init sclp_cmd_sync_early(sclp_cmdw_t cmd, void *sccb)
+{
+	int rc;
+
+	__ctl_set_bit(0, 9);
+	rc = sclp_service_call(cmd, sccb);
+	if (rc)
+		goto out;
+	__load_psw_mask(PSW_BASE_BITS | PSW_MASK_EXT |
+			PSW_MASK_WAIT | PSW_DEFAULT_KEY);
+	local_irq_disable();
+out:
+	/* Contents of the sccb might have changed. */
+	barrier();
+	__ctl_clear_bit(0, 9);
+	return rc;
+}
+
+static void __init sclp_read_info_early(void)
+{
+	int rc;
+	int i;
+	struct read_info_sccb *sccb;
+	sclp_cmdw_t commands[] = {SCLP_CMDW_READ_SCP_INFO_FORCED,
+				  SCLP_CMDW_READ_SCP_INFO};
+
+	sccb = &early_read_info_sccb;
+	for (i = 0; i < ARRAY_SIZE(commands); i++) {
+		do {
+			memset(sccb, 0, sizeof(*sccb));
+			sccb->header.length = sizeof(*sccb);
+			sccb->header.control_mask[2] = 0x80;
+			rc = sclp_cmd_sync_early(commands[i], sccb);
+		} while (rc == -EBUSY);
+
+		if (rc)
+			break;
+		if (sccb->header.response_code == 0x10) {
+			early_read_info_sccb_valid = 1;
+			break;
+		}
+		if (sccb->header.response_code != 0x1f0)
+			break;
+	}
+}
+
+void __init sclp_facilities_detect(void)
+{
+	struct read_info_sccb *sccb;
+
+	sclp_read_info_early();
+	if (!early_read_info_sccb_valid)
+		return;
+
+	sccb = &early_read_info_sccb;
+	sclp_facilities = sccb->facilities;
+	sclp_fac84 = sccb->fac84;
+	rnmax = sccb->rnmax ? sccb->rnmax : sccb->rnmax2;
+	rzm = sccb->rnsize ? sccb->rnsize : sccb->rnsize2;
+	rzm <<= 20;
+}
+
+unsigned long long sclp_get_rnmax(void)
+{
+	return rnmax;
+}
+
+unsigned long long sclp_get_rzm(void)
+{
+	return rzm;
+}
+
+/*
+ * This function will be called after sclp_facilities_detect(), which gets
+ * called from early.c code. Therefore the sccb should have valid contents.
+ */
+void __init sclp_get_ipl_info(struct sclp_ipl_info *info)
+{
+	struct read_info_sccb *sccb;
+
+	if (!early_read_info_sccb_valid)
+		return;
+	sccb = &early_read_info_sccb;
+	info->is_valid = 1;
+	if (sccb->flags & 0x2)
+		info->has_dump = 1;
+	memcpy(&info->loadparm, &sccb->loadparm, LOADPARM_LEN);
+}
+
+static void sclp_sync_callback(struct sclp_req *req, void *data)
+{
+	struct completion *completion = data;
+
+	complete(completion);
+}
+
+static int do_sync_request(sclp_cmdw_t cmd, void *sccb)
+{
+	struct completion completion;
+	struct sclp_req *request;
+	int rc;
+
+	request = kzalloc(sizeof(*request), GFP_KERNEL);
+	if (!request)
+		return -ENOMEM;
+	request->command = cmd;
+	request->sccb = sccb;
+	request->status = SCLP_REQ_FILLED;
+	request->callback = sclp_sync_callback;
+	request->callback_data = &completion;
+	init_completion(&completion);
+
+	/* Perform sclp request. */
+	rc = sclp_add_request(request);
+	if (rc)
+		goto out;
+	wait_for_completion(&completion);
+
+	/* Check response. */
+	if (request->status != SCLP_REQ_DONE) {
+		pr_warning("sync request failed (cmd=0x%08x, "
+			   "status=0x%02x)\n", cmd, request->status);
+		rc = -EIO;
+	}
+out:
+	kfree(request);
+	return rc;
+}
+
+/*
+ * CPU configuration related functions.
+ */
+
+#define SCLP_CMDW_READ_CPU_INFO		0x00010001
+#define SCLP_CMDW_CONFIGURE_CPU		0x00110001
+#define SCLP_CMDW_DECONFIGURE_CPU	0x00100001
+
+struct read_cpu_info_sccb {
+	struct	sccb_header header;
+	u16	nr_configured;
+	u16	offset_configured;
+	u16	nr_standby;
+	u16	offset_standby;
+	u8	reserved[4096 - 16];
+} __attribute__((packed, aligned(PAGE_SIZE)));
+
+static void sclp_fill_cpu_info(struct sclp_cpu_info *info,
+			       struct read_cpu_info_sccb *sccb)
+{
+	char *page = (char *) sccb;
+
+	memset(info, 0, sizeof(*info));
+	info->configured = sccb->nr_configured;
+	info->standby = sccb->nr_standby;
+	info->combined = sccb->nr_configured + sccb->nr_standby;
+	info->has_cpu_type = sclp_fac84 & 0x1;
+	memcpy(&info->cpu, page + sccb->offset_configured,
+	       info->combined * sizeof(struct sclp_cpu_entry));
+}
+
+int sclp_get_cpu_info(struct sclp_cpu_info *info)
+{
+	int rc;
+	struct read_cpu_info_sccb *sccb;
+
+	if (!SCLP_HAS_CPU_INFO)
+		return -EOPNOTSUPP;
+	sccb = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sccb)
+		return -ENOMEM;
+	sccb->header.length = sizeof(*sccb);
+	rc = do_sync_request(SCLP_CMDW_READ_CPU_INFO, sccb);
+	if (rc)
+		goto out;
+	if (sccb->header.response_code != 0x0010) {
+		pr_warning("readcpuinfo failed (response=0x%04x)\n",
+			   sccb->header.response_code);
+		rc = -EIO;
+		goto out;
+	}
+	sclp_fill_cpu_info(info, sccb);
+out:
+	free_page((unsigned long) sccb);
+	return rc;
+}
+
+struct cpu_configure_sccb {
+	struct sccb_header header;
+} __attribute__((packed, aligned(8)));
+
+static int do_cpu_configure(sclp_cmdw_t cmd)
+{
+	struct cpu_configure_sccb *sccb;
+	int rc;
+
+	if (!SCLP_HAS_CPU_RECONFIG)
+		return -EOPNOTSUPP;
+	/*
+	 * This is not going to cross a page boundary since we force
+	 * kmalloc to have a minimum alignment of 8 bytes on s390.
+	 */
+	sccb = kzalloc(sizeof(*sccb), GFP_KERNEL | GFP_DMA);
+	if (!sccb)
+		return -ENOMEM;
+	sccb->header.length = sizeof(*sccb);
+	rc = do_sync_request(cmd, sccb);
+	if (rc)
+		goto out;
+	switch (sccb->header.response_code) {
+	case 0x0020:
+	case 0x0120:
+		break;
+	default:
+		pr_warning("configure cpu failed (cmd=0x%08x, "
+			   "response=0x%04x)\n", cmd,
+			   sccb->header.response_code);
+		rc = -EIO;
+		break;
+	}
+out:
+	kfree(sccb);
+	return rc;
+}
+
+int sclp_cpu_configure(u8 cpu)
+{
+	return do_cpu_configure(SCLP_CMDW_CONFIGURE_CPU | cpu << 8);
+}
+
+int sclp_cpu_deconfigure(u8 cpu)
+{
+	return do_cpu_configure(SCLP_CMDW_DECONFIGURE_CPU | cpu << 8);
+}
+
+#ifdef CONFIG_MEMORY_HOTPLUG
+
+static DEFINE_MUTEX(sclp_mem_mutex);
+static LIST_HEAD(sclp_mem_list);
+static u8 sclp_max_storage_id;
+static unsigned long sclp_storage_ids[256 / BITS_PER_LONG];
+static int sclp_mem_state_changed;
+
+struct memory_increment {
+	struct list_head list;
+	u16 rn;
+	int standby;
+	int usecount;
+};
+
+struct assign_storage_sccb {
+	struct sccb_header header;
+	u16 rn;
+} __packed;
+
+static unsigned long long rn2addr(u16 rn)
+{
+	return (unsigned long long) (rn - 1) * rzm;
+}
+
+static int do_assign_storage(sclp_cmdw_t cmd, u16 rn)
+{
+	struct assign_storage_sccb *sccb;
+	int rc;
+
+	sccb = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sccb)
+		return -ENOMEM;
+	sccb->header.length = PAGE_SIZE;
+	sccb->rn = rn;
+	rc = do_sync_request(cmd, sccb);
+	if (rc)
+		goto out;
+	switch (sccb->header.response_code) {
+	case 0x0020:
+	case 0x0120:
+		break;
+	default:
+		pr_warning("assign storage failed (cmd=0x%08x, "
+			   "response=0x%04x, rn=0x%04x)\n", cmd,
+			   sccb->header.response_code, rn);
+		rc = -EIO;
+		break;
+	}
+out:
+	free_page((unsigned long) sccb);
+	return rc;
+}
+
+static int sclp_assign_storage(u16 rn)
+{
+	return do_assign_storage(0x000d0001, rn);
+}
+
+static int sclp_unassign_storage(u16 rn)
+{
+	return do_assign_storage(0x000c0001, rn);
+}
+
+struct attach_storage_sccb {
+	struct sccb_header header;
+	u16 :16;
+	u16 assigned;
+	u32 :32;
+	u32 entries[0];
+} __packed;
+
+static int sclp_attach_storage(u8 id)
+{
+	struct attach_storage_sccb *sccb;
+	int rc;
+	int i;
+
+	sccb = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sccb)
+		return -ENOMEM;
+	sccb->header.length = PAGE_SIZE;
+	rc = do_sync_request(0x00080001 | id << 8, sccb);
+	if (rc)
+		goto out;
+	switch (sccb->header.response_code) {
+	case 0x0020:
+		set_bit(id, sclp_storage_ids);
+		for (i = 0; i < sccb->assigned; i++)
+			sclp_unassign_storage(sccb->entries[i] >> 16);
+		break;
+	default:
+		rc = -EIO;
+		break;
+	}
+out:
+	free_page((unsigned long) sccb);
+	return rc;
+}
+
+static int sclp_mem_change_state(unsigned long start, unsigned long size,
+				 int online)
+{
+	struct memory_increment *incr;
+	unsigned long long istart;
+	int rc = 0;
+
+	list_for_each_entry(incr, &sclp_mem_list, list) {
+		istart = rn2addr(incr->rn);
+		if (start + size - 1 < istart)
+			break;
+		if (start > istart + rzm - 1)
+			continue;
+		if (online) {
+			if (incr->usecount++)
+				continue;
+			/*
+			 * Don't break the loop if one assign fails. Loop may
+			 * be walked again on CANCEL and we can't save
+			 * information if state changed before or not.
+			 * So continue and increase usecount for all increments.
+			 */
+			rc |= sclp_assign_storage(incr->rn);
+		} else {
+			if (--incr->usecount)
+				continue;
+			sclp_unassign_storage(incr->rn);
+		}
+	}
+	return rc ? -EIO : 0;
+}
+
+static int sclp_mem_notifier(struct notifier_block *nb,
+			     unsigned long action, void *data)
+{
+	unsigned long start, size;
+	struct memory_notify *arg;
+	unsigned char id;
+	int rc = 0;
+
+	arg = data;
+	start = arg->start_pfn << PAGE_SHIFT;
+	size = arg->nr_pages << PAGE_SHIFT;
+	mutex_lock(&sclp_mem_mutex);
+	for (id = 0; id <= sclp_max_storage_id; id++)
+		if (!test_bit(id, sclp_storage_ids))
+			sclp_attach_storage(id);
+	switch (action) {
+	case MEM_ONLINE:
+	case MEM_GOING_OFFLINE:
+	case MEM_CANCEL_OFFLINE:
+		break;
+	case MEM_GOING_ONLINE:
+		rc = sclp_mem_change_state(start, size, 1);
+		break;
+	case MEM_CANCEL_ONLINE:
+		sclp_mem_change_state(start, size, 0);
+		break;
+	case MEM_OFFLINE:
+		sclp_mem_change_state(start, size, 0);
+		break;
+	default:
+		rc = -EINVAL;
+		break;
+	}
+	if (!rc)
+		sclp_mem_state_changed = 1;
+	mutex_unlock(&sclp_mem_mutex);
+	return rc ? NOTIFY_BAD : NOTIFY_OK;
+}
+
+static struct notifier_block sclp_mem_nb = {
+	.notifier_call = sclp_mem_notifier,
+};
+
+static void __init add_memory_merged(u16 rn)
+{
+	static u16 first_rn, num;
+	unsigned long long start, size;
+
+	if (rn && first_rn && (first_rn + num == rn)) {
+		num++;
+		return;
+	}
+	if (!first_rn)
+		goto skip_add;
+	start = rn2addr(first_rn);
+	size = (unsigned long long ) num * rzm;
+	if (start >= VMEM_MAX_PHYS)
+		goto skip_add;
+	if (start + size > VMEM_MAX_PHYS)
+		size = VMEM_MAX_PHYS - start;
+	if (memory_end_set && (start >= memory_end))
+		goto skip_add;
+	if (memory_end_set && (start + size > memory_end))
+		size = memory_end - start;
+	add_memory(0, start, size);
+skip_add:
+	first_rn = rn;
+	num = 1;
+}
+
+static void __init sclp_add_standby_memory(void)
+{
+	struct memory_increment *incr;
+
+	list_for_each_entry(incr, &sclp_mem_list, list)
+		if (incr->standby)
+			add_memory_merged(incr->rn);
+	add_memory_merged(0);
+}
+
+static void __init insert_increment(u16 rn, int standby, int assigned)
+{
+	struct memory_increment *incr, *new_incr;
+	struct list_head *prev;
+	u16 last_rn;
+
+	new_incr = kzalloc(sizeof(*new_incr), GFP_KERNEL);
+	if (!new_incr)
+		return;
+	new_incr->rn = rn;
+	new_incr->standby = standby;
+	last_rn = 0;
+	prev = &sclp_mem_list;
+	list_for_each_entry(incr, &sclp_mem_list, list) {
+		if (assigned && incr->rn > rn)
+			break;
+		if (!assigned && incr->rn - last_rn > 1)
+			break;
+		last_rn = incr->rn;
+		prev = &incr->list;
+	}
+	if (!assigned)
+		new_incr->rn = last_rn + 1;
+	if (new_incr->rn > rnmax) {
+		kfree(new_incr);
+		return;
+	}
+	list_add(&new_incr->list, prev);
+}
+
+static int sclp_mem_freeze(struct device *dev)
+{
+	if (!sclp_mem_state_changed)
+		return 0;
+	pr_err("Memory hotplug state changed, suspend refused.\n");
+	return -EPERM;
+}
+
+struct read_storage_sccb {
+	struct sccb_header header;
+	u16 max_id;
+	u16 assigned;
+	u16 standby;
+	u16 :16;
+	u32 entries[0];
+} __packed;
+
+static struct dev_pm_ops sclp_mem_pm_ops = {
+	.freeze		= sclp_mem_freeze,
+};
+
+static struct platform_driver sclp_mem_pdrv = {
+	.driver = {
+		.name	= "sclp_mem",
+		.pm	= &sclp_mem_pm_ops,
+	},
+};
+
+static int __init sclp_detect_standby_memory(void)
+{
+	struct platform_device *sclp_pdev;
+	struct read_storage_sccb *sccb;
+	int i, id, assigned, rc;
+
+	if (!early_read_info_sccb_valid)
+		return 0;
+	if ((sclp_facilities & 0xe00000000000ULL) != 0xe00000000000ULL)
+		return 0;
+	rc = -ENOMEM;
+	sccb = (void *) __get_free_page(GFP_KERNEL | GFP_DMA);
+	if (!sccb)
+		goto out;
+	assigned = 0;
+	for (id = 0; id <= sclp_max_storage_id; id++) {
+		memset(sccb, 0, PAGE_SIZE);
+		sccb->header.length = PAGE_SIZE;
+		rc = do_sync_request(0x00040001 | id << 8, sccb);
+		if (rc)
+			goto out;
+		switch (sccb->header.response_code) {
+		case 0x0010:
+			set_bit(id, sclp_storage_ids);
+			for (i = 0; i < sccb->assigned; i++) {
+				if (!sccb->entries[i])
+					continue;
+				assigned++;
+				insert_increment(sccb->entries[i] >> 16, 0, 1);
+			}
+			break;
+		case 0x0310:
+			break;
+		case 0x0410:
+			for (i = 0; i < sccb->assigned; i++) {
+				if (!sccb->entries[i])
+					continue;
+				assigned++;
+				insert_increment(sccb->entries[i] >> 16, 1, 1);
+			}
+			break;
+		default:
+			rc = -EIO;
+			break;
+		}
+		if (!rc)
+			sclp_max_storage_id = sccb->max_id;
+	}
+	if (rc || list_empty(&sclp_mem_list))
+		goto out;
+	for (i = 1; i <= rnmax - assigned; i++)
+		insert_increment(0, 1, 0);
+	rc = register_memory_notifier(&sclp_mem_nb);
+	if (rc)
+		goto out;
+	rc = platform_driver_register(&sclp_mem_pdrv);
+	if (rc)
+		goto out;
+	sclp_pdev = platform_device_register_simple("sclp_mem", -1, NULL, 0);
+	rc = IS_ERR(sclp_pdev) ? PTR_ERR(sclp_pdev) : 0;
+	if (rc)
+		goto out_driver;
+	sclp_add_standby_memory();
+	goto out;
+out_driver:
+	platform_driver_unregister(&sclp_mem_pdrv);
+out:
+	free_page((unsigned long) sccb);
+	return rc;
+}
+__initcall(sclp_detect_standby_memory);
+
+#endif /* CONFIG_MEMORY_HOTPLUG */
+
+/*
+ * Channel path configuration related functions.
+ */
+
+#define SCLP_CMDW_CONFIGURE_CHPATH		0x000f0001
+#define SCLP_CMDW_DECONFIGURE_CHPATH		0x000e0001
+#define SCLP_CMDW_READ_CHPATH_INFORMATION	0x00030001
+
+struct chp_cfg_sccb {
+	struct sccb_header header;
+	u8 ccm;
+	u8 reserved[6];
+	u8 cssid;
+} __attribute__((packed));
+
+static int do_chp_configure(sclp_cmdw_t cmd)
+{
+	struct chp_cfg_sccb *sccb;
+	int rc;
+
+	if (!SCLP_HAS_CHP_RECONFIG)
+		return -EOPNOTSUPP;
+	/* Prepare sccb. */
+	sccb = (struct chp_cfg_sccb *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sccb)
+		return -ENOMEM;
+	sccb->header.length = sizeof(*sccb);
+	rc = do_sync_request(cmd, sccb);
+	if (rc)
+		goto out;
+	switch (sccb->header.response_code) {
+	case 0x0020:
+	case 0x0120:
+	case 0x0440:
+	case 0x0450:
+		break;
+	default:
+		pr_warning("configure channel-path failed "
+			   "(cmd=0x%08x, response=0x%04x)\n", cmd,
+			   sccb->header.response_code);
+		rc = -EIO;
+		break;
+	}
+out:
+	free_page((unsigned long) sccb);
+	return rc;
+}
+
+/**
+ * sclp_chp_configure - perform configure channel-path sclp command
+ * @chpid: channel-path ID
+ *
+ * Perform configure channel-path command sclp command for specified chpid.
+ * Return 0 after command successfully finished, non-zero otherwise.
+ */
+int sclp_chp_configure(struct chp_id chpid)
+{
+	return do_chp_configure(SCLP_CMDW_CONFIGURE_CHPATH | chpid.id << 8);
+}
+
+/**
+ * sclp_chp_deconfigure - perform deconfigure channel-path sclp command
+ * @chpid: channel-path ID
+ *
+ * Perform deconfigure channel-path command sclp command for specified chpid
+ * and wait for completion. On success return 0. Return non-zero otherwise.
+ */
+int sclp_chp_deconfigure(struct chp_id chpid)
+{
+	return do_chp_configure(SCLP_CMDW_DECONFIGURE_CHPATH | chpid.id << 8);
+}
+
+struct chp_info_sccb {
+	struct sccb_header header;
+	u8 recognized[SCLP_CHP_INFO_MASK_SIZE];
+	u8 standby[SCLP_CHP_INFO_MASK_SIZE];
+	u8 configured[SCLP_CHP_INFO_MASK_SIZE];
+	u8 ccm;
+	u8 reserved[6];
+	u8 cssid;
+} __attribute__((packed));
+
+/**
+ * sclp_chp_read_info - perform read channel-path information sclp command
+ * @info: resulting channel-path information data
+ *
+ * Perform read channel-path information sclp command and wait for completion.
+ * On success, store channel-path information in @info and return 0. Return
+ * non-zero otherwise.
+ */
+int sclp_chp_read_info(struct sclp_chp_info *info)
+{
+	struct chp_info_sccb *sccb;
+	int rc;
+
+	if (!SCLP_HAS_CHP_INFO)
+		return -EOPNOTSUPP;
+	/* Prepare sccb. */
+	sccb = (struct chp_info_sccb *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sccb)
+		return -ENOMEM;
+	sccb->header.length = sizeof(*sccb);
+	rc = do_sync_request(SCLP_CMDW_READ_CHPATH_INFORMATION, sccb);
+	if (rc)
+		goto out;
+	if (sccb->header.response_code != 0x0010) {
+		pr_warning("read channel-path info failed "
+			   "(response=0x%04x)\n", sccb->header.response_code);
+		rc = -EIO;
+		goto out;
+	}
+	memcpy(info->recognized, sccb->recognized, SCLP_CHP_INFO_MASK_SIZE);
+	memcpy(info->standby, sccb->standby, SCLP_CHP_INFO_MASK_SIZE);
+	memcpy(info->configured, sccb->configured, SCLP_CHP_INFO_MASK_SIZE);
+out:
+	free_page((unsigned long) sccb);
+	return rc;
+}

kernel/drivers/s390/char/sclp_con.c

@@ -0,0 +1,326 @@
+/*
+ * SCLP line mode console driver
+ *
+ * Copyright IBM Corp. 1999, 2009
+ * Author(s): Martin Peschke <mpeschke@de.ibm.com>
+ *	      Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#include <linux/kmod.h>
+#include <linux/console.h>
+#include <linux/init.h>
+#include <linux/timer.h>
+#include <linux/jiffies.h>
+#include <linux/termios.h>
+#include <linux/err.h>
+#include <linux/reboot.h>
+
+#include "sclp.h"
+#include "sclp_rw.h"
+#include "sclp_tty.h"
+
+#define sclp_console_major 4		/* TTYAUX_MAJOR */
+#define sclp_console_minor 64
+#define sclp_console_name  "ttyS"
+
+/* Lock to guard over changes to global variables */
+static spinlock_t sclp_con_lock;
+/* List of free pages that can be used for console output buffering */
+static struct list_head sclp_con_pages;
+/* List of full struct sclp_buffer structures ready for output */
+static struct list_head sclp_con_outqueue;
+/* Pointer to current console buffer */
+static struct sclp_buffer *sclp_conbuf;
+/* Timer for delayed output of console messages */
+static struct timer_list sclp_con_timer;
+/* Suspend mode flag */
+static int sclp_con_suspended;
+/* Flag that output queue is currently running */
+static int sclp_con_queue_running;
+
+/* Output format for console messages */
+static unsigned short sclp_con_columns;
+static unsigned short sclp_con_width_htab;
+
+static void
+sclp_conbuf_callback(struct sclp_buffer *buffer, int rc)
+{
+	unsigned long flags;
+	void *page;
+
+	do {
+		page = sclp_unmake_buffer(buffer);
+		spin_lock_irqsave(&sclp_con_lock, flags);
+
+		/* Remove buffer from outqueue */
+		list_del(&buffer->list);
+		list_add_tail((struct list_head *) page, &sclp_con_pages);
+
+		/* Check if there is a pending buffer on the out queue. */
+		buffer = NULL;
+		if (!list_empty(&sclp_con_outqueue))
+			buffer = list_first_entry(&sclp_con_outqueue,
+						  struct sclp_buffer, list);
+		if (!buffer || sclp_con_suspended) {
+			sclp_con_queue_running = 0;
+			spin_unlock_irqrestore(&sclp_con_lock, flags);
+			break;
+		}
+		spin_unlock_irqrestore(&sclp_con_lock, flags);
+	} while (sclp_emit_buffer(buffer, sclp_conbuf_callback));
+}
+
+/*
+ * Finalize and emit first pending buffer.
+ */
+static void sclp_conbuf_emit(void)
+{
+	struct sclp_buffer* buffer;
+	unsigned long flags;
+	int rc;
+
+	spin_lock_irqsave(&sclp_con_lock, flags);
+	if (sclp_conbuf)
+		list_add_tail(&sclp_conbuf->list, &sclp_con_outqueue);
+	sclp_conbuf = NULL;
+	if (sclp_con_queue_running || sclp_con_suspended)
+		goto out_unlock;
+	if (list_empty(&sclp_con_outqueue))
+		goto out_unlock;
+	buffer = list_first_entry(&sclp_con_outqueue, struct sclp_buffer,
+				  list);
+	sclp_con_queue_running = 1;
+	spin_unlock_irqrestore(&sclp_con_lock, flags);
+
+	rc = sclp_emit_buffer(buffer, sclp_conbuf_callback);
+	if (rc)
+		sclp_conbuf_callback(buffer, rc);
+	return;
+out_unlock:
+	spin_unlock_irqrestore(&sclp_con_lock, flags);
+}
+
+/*
+ * Wait until out queue is empty
+ */
+static void sclp_console_sync_queue(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sclp_con_lock, flags);
+	if (timer_pending(&sclp_con_timer))
+		del_timer(&sclp_con_timer);
+	while (sclp_con_queue_running) {
+		spin_unlock_irqrestore(&sclp_con_lock, flags);
+		sclp_sync_wait();
+		spin_lock_irqsave(&sclp_con_lock, flags);
+	}
+	spin_unlock_irqrestore(&sclp_con_lock, flags);
+}
+
+/*
+ * When this routine is called from the timer then we flush the
+ * temporary write buffer without further waiting on a final new line.
+ */
+static void
+sclp_console_timeout(unsigned long data)
+{
+	sclp_conbuf_emit();
+}
+
+/*
+ * Writes the given message to S390 system console
+ */
+static void
+sclp_console_write(struct console *console, const char *message,
+		   unsigned int count)
+{
+	unsigned long flags;
+	void *page;
+	int written;
+
+	if (count == 0)
+		return;
+	spin_lock_irqsave(&sclp_con_lock, flags);
+	/*
+	 * process escape characters, write message into buffer,
+	 * send buffer to SCLP
+	 */
+	do {
+		/* make sure we have a console output buffer */
+		if (sclp_conbuf == NULL) {
+			while (list_empty(&sclp_con_pages)) {
+				if (sclp_con_suspended)
+					goto out;
+				spin_unlock_irqrestore(&sclp_con_lock, flags);
+				sclp_sync_wait();
+				spin_lock_irqsave(&sclp_con_lock, flags);
+			}
+			page = sclp_con_pages.next;
+			list_del((struct list_head *) page);
+			sclp_conbuf = sclp_make_buffer(page, sclp_con_columns,
+						       sclp_con_width_htab);
+		}
+		/* try to write the string to the current output buffer */
+		written = sclp_write(sclp_conbuf, (const unsigned char *)
+				     message, count);
+		if (written == count)
+			break;
+		/*
+		 * Not all characters could be written to the current
+		 * output buffer. Emit the buffer, create a new buffer
+		 * and then output the rest of the string.
+		 */
+		spin_unlock_irqrestore(&sclp_con_lock, flags);
+		sclp_conbuf_emit();
+		spin_lock_irqsave(&sclp_con_lock, flags);
+		message += written;
+		count -= written;
+	} while (count > 0);
+	/* Setup timer to output current console buffer after 1/10 second */
+	if (sclp_conbuf != NULL && sclp_chars_in_buffer(sclp_conbuf) != 0 &&
+	    !timer_pending(&sclp_con_timer)) {
+		init_timer(&sclp_con_timer);
+		sclp_con_timer.function = sclp_console_timeout;
+		sclp_con_timer.data = 0UL;
+		sclp_con_timer.expires = jiffies + HZ/10;
+		add_timer(&sclp_con_timer);
+	}
+out:
+	spin_unlock_irqrestore(&sclp_con_lock, flags);
+}
+
+static struct tty_driver *
+sclp_console_device(struct console *c, int *index)
+{
+	*index = c->index;
+	return sclp_tty_driver;
+}
+
+/*
+ * Make sure that all buffers will be flushed to the SCLP.
+ */
+static void
+sclp_console_flush(void)
+{
+	sclp_conbuf_emit();
+	sclp_console_sync_queue();
+}
+
+/*
+ * Resume console: If there are cached messages, emit them.
+ */
+static void sclp_console_resume(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sclp_con_lock, flags);
+	sclp_con_suspended = 0;
+	spin_unlock_irqrestore(&sclp_con_lock, flags);
+	sclp_conbuf_emit();
+}
+
+/*
+ * Suspend console: Set suspend flag and flush console
+ */
+static void sclp_console_suspend(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sclp_con_lock, flags);
+	sclp_con_suspended = 1;
+	spin_unlock_irqrestore(&sclp_con_lock, flags);
+	sclp_console_flush();
+}
+
+static int sclp_console_notify(struct notifier_block *self,
+			       unsigned long event, void *data)
+{
+	sclp_console_flush();
+	return NOTIFY_OK;
+}
+
+static struct notifier_block on_panic_nb = {
+	.notifier_call = sclp_console_notify,
+	.priority = SCLP_PANIC_PRIO_CLIENT,
+};
+
+static struct notifier_block on_reboot_nb = {
+	.notifier_call = sclp_console_notify,
+	.priority = 1,
+};
+
+/*
+ * used to register the SCLP console to the kernel and to
+ * give printk necessary information
+ */
+static struct console sclp_console =
+{
+	.name = sclp_console_name,
+	.write = sclp_console_write,
+	.device = sclp_console_device,
+	.flags = CON_PRINTBUFFER,
+	.index = 0 /* ttyS0 */
+};
+
+/*
+ * This function is called for SCLP suspend and resume events.
+ */
+void sclp_console_pm_event(enum sclp_pm_event sclp_pm_event)
+{
+	switch (sclp_pm_event) {
+	case SCLP_PM_EVENT_FREEZE:
+		sclp_console_suspend();
+		break;
+	case SCLP_PM_EVENT_RESTORE:
+	case SCLP_PM_EVENT_THAW:
+		sclp_console_resume();
+		break;
+	}
+}
+
+/*
+ * called by console_init() in drivers/char/tty_io.c at boot-time.
+ */
+static int __init
+sclp_console_init(void)
+{
+	void *page;
+	int i;
+	int rc;
+
+	if (!CONSOLE_IS_SCLP)
+		return 0;
+	rc = sclp_rw_init();
+	if (rc)
+		return rc;
+	/* Allocate pages for output buffering */
+	INIT_LIST_HEAD(&sclp_con_pages);
+	for (i = 0; i < MAX_CONSOLE_PAGES; i++) {
+		page = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		list_add_tail(page, &sclp_con_pages);
+	}
+	INIT_LIST_HEAD(&sclp_con_outqueue);
+	spin_lock_init(&sclp_con_lock);
+	sclp_conbuf = NULL;
+	init_timer(&sclp_con_timer);
+
+	/* Set output format */
+	if (MACHINE_IS_VM)
+		/*
+		 * save 4 characters for the CPU number
+		 * written at start of each line by VM/CP
+		 */
+		sclp_con_columns = 76;
+	else
+		sclp_con_columns = 80;
+	sclp_con_width_htab = 8;
+
+	/* enable printk-access to this driver */
+	atomic_notifier_chain_register(&panic_notifier_list, &on_panic_nb);
+	register_reboot_notifier(&on_reboot_nb);
+	register_console(&sclp_console);
+	return 0;
+}
+
+console_initcall(sclp_console_init);

kernel/drivers/s390/char/sclp_config.c

@@ -0,0 +1,90 @@
+/*
+ *  drivers/s390/char/sclp_config.c
+ *
+ *    Copyright IBM Corp. 2007
+ *    Author(s): Heiko Carstens <heiko.carstens@de.ibm.com>
+ */
+
+#define KMSG_COMPONENT "sclp_config"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/cpu.h>
+#include <linux/sysdev.h>
+#include <linux/workqueue.h>
+#include <asm/smp.h>
+
+#include "sclp.h"
+
+struct conf_mgm_data {
+	u8 reserved;
+	u8 ev_qualifier;
+} __attribute__((packed));
+
+#define EV_QUAL_CPU_CHANGE	1
+#define EV_QUAL_CAP_CHANGE	3
+
+static struct work_struct sclp_cpu_capability_work;
+static struct work_struct sclp_cpu_change_work;
+
+static void sclp_cpu_capability_notify(struct work_struct *work)
+{
+	int cpu;
+	struct sys_device *sysdev;
+
+	pr_warning("cpu capability changed.\n");
+	get_online_cpus();
+	for_each_online_cpu(cpu) {
+		sysdev = get_cpu_sysdev(cpu);
+		kobject_uevent(&sysdev->kobj, KOBJ_CHANGE);
+	}
+	put_online_cpus();
+}
+
+static void __ref sclp_cpu_change_notify(struct work_struct *work)
+{
+	smp_rescan_cpus();
+}
+
+static void sclp_conf_receiver_fn(struct evbuf_header *evbuf)
+{
+	struct conf_mgm_data *cdata;
+
+	cdata = (struct conf_mgm_data *)(evbuf + 1);
+	switch (cdata->ev_qualifier) {
+	case EV_QUAL_CPU_CHANGE:
+		schedule_work(&sclp_cpu_change_work);
+		break;
+	case EV_QUAL_CAP_CHANGE:
+		schedule_work(&sclp_cpu_capability_work);
+		break;
+	}
+}
+
+static struct sclp_register sclp_conf_register =
+{
+	.receive_mask = EVTYP_CONFMGMDATA_MASK,
+	.receiver_fn  = sclp_conf_receiver_fn,
+};
+
+static int __init sclp_conf_init(void)
+{
+	int rc;
+
+	INIT_WORK(&sclp_cpu_capability_work, sclp_cpu_capability_notify);
+	INIT_WORK(&sclp_cpu_change_work, sclp_cpu_change_notify);
+
+	rc = sclp_register(&sclp_conf_register);
+	if (rc)
+		return rc;
+
+	if (!(sclp_conf_register.sclp_send_mask & EVTYP_CONFMGMDATA_MASK)) {
+		pr_warning("no configuration management.\n");
+		sclp_unregister(&sclp_conf_register);
+		rc = -ENOSYS;
+	}
+	return rc;
+}
+
+__initcall(sclp_conf_init);

kernel/drivers/s390/char/sclp_cpi.c

@@ -0,0 +1,41 @@
+/*
+ *  drivers/s390/char/sclp_cpi.c
+ *    SCLP control programm identification
+ *
+ *    Copyright IBM Corp. 2001, 2007
+ *    Author(s): Martin Peschke <mpeschke@de.ibm.com>
+ *		 Michael Ernst <mernst@de.ibm.com>
+ */
+
+#include <linux/kmod.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/version.h>
+#include "sclp_cpi_sys.h"
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Identify this operating system instance "
+		   "to the System z hardware");
+MODULE_AUTHOR("Martin Peschke <mpeschke@de.ibm.com>, "
+	      "Michael Ernst <mernst@de.ibm.com>");
+
+static char *system_name = "";
+static char *sysplex_name = "";
+
+module_param(system_name, charp, 0);
+MODULE_PARM_DESC(system_name, "e.g. hostname - max. 8 characters");
+module_param(sysplex_name, charp, 0);
+MODULE_PARM_DESC(sysplex_name, "if applicable - max. 8 characters");
+
+static int __init cpi_module_init(void)
+{
+	return sclp_cpi_set_data(system_name, sysplex_name, "LINUX",
+				 LINUX_VERSION_CODE);
+}
+
+static void __exit cpi_module_exit(void)
+{
+}
+
+module_init(cpi_module_init);
+module_exit(cpi_module_exit);

kernel/drivers/s390/char/sclp_cpi_sys.c

@@ -0,0 +1,429 @@
+/*
+ *  drivers/s390/char/sclp_cpi_sys.c
+ *    SCLP control program identification sysfs interface
+ *
+ *    Copyright IBM Corp. 2001, 2007
+ *    Author(s): Martin Peschke <mpeschke@de.ibm.com>
+ *		 Michael Ernst <mernst@de.ibm.com>
+ */
+
+#define KMSG_COMPONENT "sclp_cpi"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/stat.h>
+#include <linux/device.h>
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/kmod.h>
+#include <linux/timer.h>
+#include <linux/err.h>
+#include <linux/slab.h>
+#include <linux/completion.h>
+#include <asm/ebcdic.h>
+#include <asm/sclp.h>
+
+#include "sclp.h"
+#include "sclp_rw.h"
+#include "sclp_cpi_sys.h"
+
+#define CPI_LENGTH_NAME 8
+#define CPI_LENGTH_LEVEL 16
+
+static DEFINE_MUTEX(sclp_cpi_mutex);
+
+struct cpi_evbuf {
+	struct evbuf_header header;
+	u8	id_format;
+	u8	reserved0;
+	u8	system_type[CPI_LENGTH_NAME];
+	u64	reserved1;
+	u8	system_name[CPI_LENGTH_NAME];
+	u64	reserved2;
+	u64	system_level;
+	u64	reserved3;
+	u8	sysplex_name[CPI_LENGTH_NAME];
+	u8	reserved4[16];
+} __attribute__((packed));
+
+struct cpi_sccb {
+	struct sccb_header header;
+	struct cpi_evbuf cpi_evbuf;
+} __attribute__((packed));
+
+static struct sclp_register sclp_cpi_event = {
+	.send_mask = EVTYP_CTLPROGIDENT_MASK,
+};
+
+static char system_name[CPI_LENGTH_NAME + 1];
+static char sysplex_name[CPI_LENGTH_NAME + 1];
+static char system_type[CPI_LENGTH_NAME + 1];
+static u64 system_level;
+
+static void set_data(char *field, char *data)
+{
+	memset(field, ' ', CPI_LENGTH_NAME);
+	memcpy(field, data, strlen(data));
+	sclp_ascebc_str(field, CPI_LENGTH_NAME);
+}
+
+static void cpi_callback(struct sclp_req *req, void *data)
+{
+	struct completion *completion = data;
+
+	complete(completion);
+}
+
+static struct sclp_req *cpi_prepare_req(void)
+{
+	struct sclp_req *req;
+	struct cpi_sccb *sccb;
+	struct cpi_evbuf *evb;
+
+	req = kzalloc(sizeof(struct sclp_req), GFP_KERNEL);
+	if (!req)
+		return ERR_PTR(-ENOMEM);
+	sccb = (struct cpi_sccb *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sccb) {
+		kfree(req);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	/* setup SCCB for Control-Program Identification */
+	sccb->header.length = sizeof(struct cpi_sccb);
+	sccb->cpi_evbuf.header.length = sizeof(struct cpi_evbuf);
+	sccb->cpi_evbuf.header.type = 0x0b;
+	evb = &sccb->cpi_evbuf;
+
+	/* set system type */
+	set_data(evb->system_type, system_type);
+
+	/* set system name */
+	set_data(evb->system_name, system_name);
+
+	/* set sytem level */
+	evb->system_level = system_level;
+
+	/* set sysplex name */
+	set_data(evb->sysplex_name, sysplex_name);
+
+	/* prepare request data structure presented to SCLP driver */
+	req->command = SCLP_CMDW_WRITE_EVENT_DATA;
+	req->sccb = sccb;
+	req->status = SCLP_REQ_FILLED;
+	req->callback = cpi_callback;
+	return req;
+}
+
+static void cpi_free_req(struct sclp_req *req)
+{
+	free_page((unsigned long) req->sccb);
+	kfree(req);
+}
+
+static int cpi_req(void)
+{
+	struct completion completion;
+	struct sclp_req *req;
+	int rc;
+	int response;
+
+	rc = sclp_register(&sclp_cpi_event);
+	if (rc)
+		goto out;
+	if (!(sclp_cpi_event.sclp_receive_mask & EVTYP_CTLPROGIDENT_MASK)) {
+		rc = -EOPNOTSUPP;
+		goto out_unregister;
+	}
+
+	req = cpi_prepare_req();
+	if (IS_ERR(req)) {
+		rc = PTR_ERR(req);
+		goto out_unregister;
+	}
+
+	init_completion(&completion);
+	req->callback_data = &completion;
+
+	/* Add request to sclp queue */
+	rc = sclp_add_request(req);
+	if (rc)
+		goto out_free_req;
+
+	wait_for_completion(&completion);
+
+	if (req->status != SCLP_REQ_DONE) {
+		pr_warning("request failed (status=0x%02x)\n",
+			   req->status);
+		rc = -EIO;
+		goto out_free_req;
+	}
+
+	response = ((struct cpi_sccb *) req->sccb)->header.response_code;
+	if (response != 0x0020) {
+		pr_warning("request failed with response code 0x%x\n",
+			   response);
+		rc = -EIO;
+	}
+
+out_free_req:
+	cpi_free_req(req);
+
+out_unregister:
+	sclp_unregister(&sclp_cpi_event);
+
+out:
+	return rc;
+}
+
+static int check_string(const char *attr, const char *str)
+{
+	size_t len;
+	size_t i;
+
+	len = strlen(str);
+
+	if ((len > 0) && (str[len - 1] == '\n'))
+		len--;
+
+	if (len > CPI_LENGTH_NAME)
+		return -EINVAL;
+
+	for (i = 0; i < len ; i++) {
+		if (isalpha(str[i]) || isdigit(str[i]) ||
+		    strchr("$@# ", str[i]))
+			continue;
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void set_string(char *attr, const char *value)
+{
+	size_t len;
+	size_t i;
+
+	len = strlen(value);
+
+	if ((len > 0) && (value[len - 1] == '\n'))
+		len--;
+
+	for (i = 0; i < CPI_LENGTH_NAME; i++) {
+		if (i < len)
+			attr[i] = toupper(value[i]);
+		else
+			attr[i] = ' ';
+	}
+}
+
+static ssize_t system_name_show(struct kobject *kobj,
+				struct kobj_attribute *attr, char *page)
+{
+	int rc;
+
+	mutex_lock(&sclp_cpi_mutex);
+	rc = snprintf(page, PAGE_SIZE, "%s\n", system_name);
+	mutex_unlock(&sclp_cpi_mutex);
+	return rc;
+}
+
+static ssize_t system_name_store(struct kobject *kobj,
+				 struct kobj_attribute *attr,
+				 const char *buf,
+	size_t len)
+{
+	int rc;
+
+	rc = check_string("system_name", buf);
+	if (rc)
+		return rc;
+
+	mutex_lock(&sclp_cpi_mutex);
+	set_string(system_name, buf);
+	mutex_unlock(&sclp_cpi_mutex);
+
+	return len;
+}
+
+static struct kobj_attribute system_name_attr =
+	__ATTR(system_name, 0644, system_name_show, system_name_store);
+
+static ssize_t sysplex_name_show(struct kobject *kobj,
+				 struct kobj_attribute *attr, char *page)
+{
+	int rc;
+
+	mutex_lock(&sclp_cpi_mutex);
+	rc = snprintf(page, PAGE_SIZE, "%s\n", sysplex_name);
+	mutex_unlock(&sclp_cpi_mutex);
+	return rc;
+}
+
+static ssize_t sysplex_name_store(struct kobject *kobj,
+				  struct kobj_attribute *attr,
+				  const char *buf,
+	size_t len)
+{
+	int rc;
+
+	rc = check_string("sysplex_name", buf);
+	if (rc)
+		return rc;
+
+	mutex_lock(&sclp_cpi_mutex);
+	set_string(sysplex_name, buf);
+	mutex_unlock(&sclp_cpi_mutex);
+
+	return len;
+}
+
+static struct kobj_attribute sysplex_name_attr =
+	__ATTR(sysplex_name, 0644, sysplex_name_show, sysplex_name_store);
+
+static ssize_t system_type_show(struct kobject *kobj,
+				struct kobj_attribute *attr, char *page)
+{
+	int rc;
+
+	mutex_lock(&sclp_cpi_mutex);
+	rc = snprintf(page, PAGE_SIZE, "%s\n", system_type);
+	mutex_unlock(&sclp_cpi_mutex);
+	return rc;
+}
+
+static ssize_t system_type_store(struct kobject *kobj,
+				 struct kobj_attribute *attr,
+				 const char *buf,
+	size_t len)
+{
+	int rc;
+
+	rc = check_string("system_type", buf);
+	if (rc)
+		return rc;
+
+	mutex_lock(&sclp_cpi_mutex);
+	set_string(system_type, buf);
+	mutex_unlock(&sclp_cpi_mutex);
+
+	return len;
+}
+
+static struct kobj_attribute system_type_attr =
+	__ATTR(system_type, 0644, system_type_show, system_type_store);
+
+static ssize_t system_level_show(struct kobject *kobj,
+				 struct kobj_attribute *attr, char *page)
+{
+	unsigned long long level;
+
+	mutex_lock(&sclp_cpi_mutex);
+	level = system_level;
+	mutex_unlock(&sclp_cpi_mutex);
+	return snprintf(page, PAGE_SIZE, "%#018llx\n", level);
+}
+
+static ssize_t system_level_store(struct kobject *kobj,
+				  struct kobj_attribute *attr,
+				  const char *buf,
+	size_t len)
+{
+	unsigned long long level;
+	char *endp;
+
+	level = simple_strtoull(buf, &endp, 16);
+
+	if (endp == buf)
+		return -EINVAL;
+	if (*endp == '\n')
+		endp++;
+	if (*endp)
+		return -EINVAL;
+
+	mutex_lock(&sclp_cpi_mutex);
+	system_level = level;
+	mutex_unlock(&sclp_cpi_mutex);
+	return len;
+}
+
+static struct kobj_attribute system_level_attr =
+	__ATTR(system_level, 0644, system_level_show, system_level_store);
+
+static ssize_t set_store(struct kobject *kobj,
+			 struct kobj_attribute *attr,
+			 const char *buf, size_t len)
+{
+	int rc;
+
+	mutex_lock(&sclp_cpi_mutex);
+	rc = cpi_req();
+	mutex_unlock(&sclp_cpi_mutex);
+	if (rc)
+		return rc;
+
+	return len;
+}
+
+static struct kobj_attribute set_attr = __ATTR(set, 0200, NULL, set_store);
+
+static struct attribute *cpi_attrs[] = {
+	&system_name_attr.attr,
+	&sysplex_name_attr.attr,
+	&system_type_attr.attr,
+	&system_level_attr.attr,
+	&set_attr.attr,
+	NULL,
+};
+
+static struct attribute_group cpi_attr_group = {
+	.attrs = cpi_attrs,
+};
+
+static struct kset *cpi_kset;
+
+int sclp_cpi_set_data(const char *system, const char *sysplex, const char *type,
+		      const u64 level)
+{
+	int rc;
+
+	rc = check_string("system_name", system);
+	if (rc)
+		return rc;
+	rc = check_string("sysplex_name", sysplex);
+	if (rc)
+		return rc;
+	rc = check_string("system_type", type);
+	if (rc)
+		return rc;
+
+	mutex_lock(&sclp_cpi_mutex);
+	set_string(system_name, system);
+	set_string(sysplex_name, sysplex);
+	set_string(system_type, type);
+	system_level = level;
+
+	rc = cpi_req();
+	mutex_unlock(&sclp_cpi_mutex);
+
+	return rc;
+}
+EXPORT_SYMBOL(sclp_cpi_set_data);
+
+static int __init cpi_init(void)
+{
+	int rc;
+
+	cpi_kset = kset_create_and_add("cpi", NULL, firmware_kobj);
+	if (!cpi_kset)
+		return -ENOMEM;
+
+	rc = sysfs_create_group(&cpi_kset->kobj, &cpi_attr_group);
+	if (rc)
+		kset_unregister(cpi_kset);
+
+	return rc;
+}
+
+__initcall(cpi_init);

kernel/drivers/s390/char/sclp_cpi_sys.h

@@ -0,0 +1,15 @@
+/*
+ *  drivers/s390/char/sclp_cpi_sys.h
+ *    SCLP control program identification sysfs interface
+ *
+ *    Copyright IBM Corp. 2007
+ *    Author(s): Michael Ernst <mernst@de.ibm.com>
+ */
+
+#ifndef __SCLP_CPI_SYS_H__
+#define __SCLP_CPI_SYS_H__
+
+int sclp_cpi_set_data(const char *system, const char *sysplex,
+		      const char *type, u64 level);
+
+#endif	 /* __SCLP_CPI_SYS_H__ */

kernel/drivers/s390/char/sclp_quiesce.c

@@ -0,0 +1,85 @@
+/*
+ *  drivers/s390/char/sclp_quiesce.c
+ *     signal quiesce handler
+ *
+ *  (C) Copyright IBM Corp. 1999,2004
+ *  Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *             Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/cpumask.h>
+#include <linux/smp.h>
+#include <linux/init.h>
+#include <linux/reboot.h>
+#include <asm/atomic.h>
+#include <asm/ptrace.h>
+#include <asm/sigp.h>
+#include <asm/smp.h>
+
+#include "sclp.h"
+
+static void (*old_machine_restart)(char *);
+static void (*old_machine_halt)(void);
+static void (*old_machine_power_off)(void);
+
+/* Shutdown handler. Signal completion of shutdown by loading special PSW. */
+static void do_machine_quiesce(void)
+{
+	psw_t quiesce_psw;
+
+	smp_send_stop();
+	quiesce_psw.mask = PSW_BASE_BITS | PSW_MASK_WAIT;
+	quiesce_psw.addr = 0xfff;
+	__load_psw(quiesce_psw);
+}
+
+/* Handler for quiesce event. Start shutdown procedure. */
+static void sclp_quiesce_handler(struct evbuf_header *evbuf)
+{
+	if (_machine_restart != (void *) do_machine_quiesce) {
+		old_machine_restart = _machine_restart;
+		old_machine_halt = _machine_halt;
+		old_machine_power_off = _machine_power_off;
+		_machine_restart = (void *) do_machine_quiesce;
+		_machine_halt = do_machine_quiesce;
+		_machine_power_off = do_machine_quiesce;
+	}
+	ctrl_alt_del();
+}
+
+/* Undo machine restart/halt/power_off modification on resume */
+static void sclp_quiesce_pm_event(struct sclp_register *reg,
+				  enum sclp_pm_event sclp_pm_event)
+{
+	switch (sclp_pm_event) {
+	case SCLP_PM_EVENT_RESTORE:
+		if (old_machine_restart) {
+			_machine_restart = old_machine_restart;
+			_machine_halt = old_machine_halt;
+			_machine_power_off = old_machine_power_off;
+			old_machine_restart = NULL;
+			old_machine_halt = NULL;
+			old_machine_power_off = NULL;
+		}
+		break;
+	case SCLP_PM_EVENT_FREEZE:
+	case SCLP_PM_EVENT_THAW:
+		break;
+	}
+}
+
+static struct sclp_register sclp_quiesce_event = {
+	.receive_mask = EVTYP_SIGQUIESCE_MASK,
+	.receiver_fn = sclp_quiesce_handler,
+	.pm_event_fn = sclp_quiesce_pm_event
+};
+
+/* Initialize quiesce driver. */
+static int __init sclp_quiesce_init(void)
+{
+	return sclp_register(&sclp_quiesce_event);
+}
+
+module_init(sclp_quiesce_init);

kernel/drivers/s390/char/sclp_rw.c

@@ -0,0 +1,474 @@
+/*
+ * driver: reading from and writing to system console on S/390 via SCLP
+ *
+ * Copyright IBM Corp. 1999, 2009
+ *
+ * Author(s): Martin Peschke <mpeschke@de.ibm.com>
+ *	      Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#include <linux/kmod.h>
+#include <linux/types.h>
+#include <linux/err.h>
+#include <linux/string.h>
+#include <linux/spinlock.h>
+#include <linux/ctype.h>
+#include <asm/uaccess.h>
+
+#include "sclp.h"
+#include "sclp_rw.h"
+
+/*
+ * The room for the SCCB (only for writing) is not equal to a pages size
+ * (as it is specified as the maximum size in the SCLP documentation)
+ * because of the additional data structure described above.
+ */
+#define MAX_SCCB_ROOM (PAGE_SIZE - sizeof(struct sclp_buffer))
+
+static void sclp_rw_pm_event(struct sclp_register *reg,
+			     enum sclp_pm_event sclp_pm_event)
+{
+	sclp_console_pm_event(sclp_pm_event);
+}
+
+/* Event type structure for write message and write priority message */
+static struct sclp_register sclp_rw_event = {
+	.send_mask = EVTYP_MSG_MASK | EVTYP_PMSGCMD_MASK,
+	.pm_event_fn = sclp_rw_pm_event,
+};
+
+/*
+ * Setup a sclp write buffer. Gets a page as input (4K) and returns
+ * a pointer to a struct sclp_buffer structure that is located at the
+ * end of the input page. This reduces the buffer space by a few
+ * bytes but simplifies things.
+ */
+struct sclp_buffer *
+sclp_make_buffer(void *page, unsigned short columns, unsigned short htab)
+{
+	struct sclp_buffer *buffer;
+	struct write_sccb *sccb;
+
+	sccb = (struct write_sccb *) page;
+	/*
+	 * We keep the struct sclp_buffer structure at the end
+	 * of the sccb page.
+	 */
+	buffer = ((struct sclp_buffer *) ((addr_t) sccb + PAGE_SIZE)) - 1;
+	buffer->sccb = sccb;
+	buffer->retry_count = 0;
+	buffer->mto_number = 0;
+	buffer->mto_char_sum = 0;
+	buffer->current_line = NULL;
+	buffer->current_length = 0;
+	buffer->columns = columns;
+	buffer->htab = htab;
+
+	/* initialize sccb */
+	memset(sccb, 0, sizeof(struct write_sccb));
+	sccb->header.length = sizeof(struct write_sccb);
+	sccb->msg_buf.header.length = sizeof(struct msg_buf);
+	sccb->msg_buf.header.type = EVTYP_MSG;
+	sccb->msg_buf.mdb.header.length = sizeof(struct mdb);
+	sccb->msg_buf.mdb.header.type = 1;
+	sccb->msg_buf.mdb.header.tag = 0xD4C4C240;	/* ebcdic "MDB " */
+	sccb->msg_buf.mdb.header.revision_code = 1;
+	sccb->msg_buf.mdb.go.length = sizeof(struct go);
+	sccb->msg_buf.mdb.go.type = 1;
+
+	return buffer;
+}
+
+/*
+ * Return a pointer to the original page that has been used to create
+ * the buffer.
+ */
+void *
+sclp_unmake_buffer(struct sclp_buffer *buffer)
+{
+	return buffer->sccb;
+}
+
+/*
+ * Initialize a new Message Text Object (MTO) at the end of the provided buffer
+ * with enough room for max_len characters. Return 0 on success.
+ */
+static int
+sclp_initialize_mto(struct sclp_buffer *buffer, int max_len)
+{
+	struct write_sccb *sccb;
+	struct mto *mto;
+	int mto_size;
+
+	/* max size of new Message Text Object including message text  */
+	mto_size = sizeof(struct mto) + max_len;
+
+	/* check if current buffer sccb can contain the mto */
+	sccb = buffer->sccb;
+	if ((MAX_SCCB_ROOM - sccb->header.length) < mto_size)
+		return -ENOMEM;
+
+	/* find address of new message text object */
+	mto = (struct mto *)(((addr_t) sccb) + sccb->header.length);
+
+	/*
+	 * fill the new Message-Text Object,
+	 * starting behind the former last byte of the SCCB
+	 */
+	memset(mto, 0, sizeof(struct mto));
+	mto->length = sizeof(struct mto);
+	mto->type = 4;	/* message text object */
+	mto->line_type_flags = LNTPFLGS_ENDTEXT; /* end text */
+
+	/* set pointer to first byte after struct mto. */
+	buffer->current_line = (char *) (mto + 1);
+	buffer->current_length = 0;
+
+	return 0;
+}
+
+/*
+ * Finalize MTO initialized by sclp_initialize_mto(), updating the sizes of
+ * MTO, enclosing MDB, event buffer and SCCB.
+ */
+static void
+sclp_finalize_mto(struct sclp_buffer *buffer)
+{
+	struct write_sccb *sccb;
+	struct mto *mto;
+	int str_len, mto_size;
+
+	str_len = buffer->current_length;
+	buffer->current_line = NULL;
+	buffer->current_length = 0;
+
+	/* real size of new Message Text Object including message text	*/
+	mto_size = sizeof(struct mto) + str_len;
+
+	/* find address of new message text object */
+	sccb = buffer->sccb;
+	mto = (struct mto *)(((addr_t) sccb) + sccb->header.length);
+
+	/* set size of message text object */
+	mto->length = mto_size;
+
+	/*
+	 * update values of sizes
+	 * (SCCB, Event(Message) Buffer, Message Data Block)
+	 */
+	sccb->header.length += mto_size;
+	sccb->msg_buf.header.length += mto_size;
+	sccb->msg_buf.mdb.header.length += mto_size;
+
+	/*
+	 * count number of buffered messages (= number of Message Text
+	 * Objects) and number of buffered characters
+	 * for the SCCB currently used for buffering and at all
+	 */
+	buffer->mto_number++;
+	buffer->mto_char_sum += str_len;
+}
+
+/*
+ * processing of a message including escape characters,
+ * returns number of characters written to the output sccb
+ * ("processed" means that is not guaranteed that the character have already
+ *  been sent to the SCLP but that it will be done at least next time the SCLP
+ *  is not busy)
+ */
+int
+sclp_write(struct sclp_buffer *buffer, const unsigned char *msg, int count)
+{
+	int spaces, i_msg;
+	int rc;
+
+	/*
+	 * parse msg for escape sequences (\t,\v ...) and put formated
+	 * msg into an mto (created by sclp_initialize_mto).
+	 *
+	 * We have to do this work ourselfs because there is no support for
+	 * these characters on the native machine and only partial support
+	 * under VM (Why does VM interpret \n but the native machine doesn't ?)
+	 *
+	 * Depending on i/o-control setting the message is always written
+	 * immediately or we wait for a final new line maybe coming with the
+	 * next message. Besides we avoid a buffer overrun by writing its
+	 * content.
+	 *
+	 * RESTRICTIONS:
+	 *
+	 * \r and \b work within one line because we are not able to modify
+	 * previous output that have already been accepted by the SCLP.
+	 *
+	 * \t combined with following \r is not correctly represented because
+	 * \t is expanded to some spaces but \r does not know about a
+	 * previous \t and decreases the current position by one column.
+	 * This is in order to a slim and quick implementation.
+	 */
+	for (i_msg = 0; i_msg < count; i_msg++) {
+		switch (msg[i_msg]) {
+		case '\n':	/* new line, line feed (ASCII)	*/
+			/* check if new mto needs to be created */
+			if (buffer->current_line == NULL) {
+				rc = sclp_initialize_mto(buffer, 0);
+				if (rc)
+					return i_msg;
+			}
+			sclp_finalize_mto(buffer);
+			break;
+		case '\a':	/* bell, one for several times	*/
+			/* set SCLP sound alarm bit in General Object */
+			buffer->sccb->msg_buf.mdb.go.general_msg_flags |=
+				GNRLMSGFLGS_SNDALRM;
+			break;
+		case '\t':	/* horizontal tabulator	 */
+			/* check if new mto needs to be created */
+			if (buffer->current_line == NULL) {
+				rc = sclp_initialize_mto(buffer,
+							 buffer->columns);
+				if (rc)
+					return i_msg;
+			}
+			/* "go to (next htab-boundary + 1, same line)" */
+			do {
+				if (buffer->current_length >= buffer->columns)
+					break;
+				/* ok, add a blank */
+				*buffer->current_line++ = 0x40;
+				buffer->current_length++;
+			} while (buffer->current_length % buffer->htab);
+			break;
+		case '\f':	/* form feed  */
+		case '\v':	/* vertical tabulator  */
+			/* "go to (actual column, actual line + 1)" */
+			/* = new line, leading spaces */
+			if (buffer->current_line != NULL) {
+				spaces = buffer->current_length;
+				sclp_finalize_mto(buffer);
+				rc = sclp_initialize_mto(buffer,
+							 buffer->columns);
+				if (rc)
+					return i_msg;
+				memset(buffer->current_line, 0x40, spaces);
+				buffer->current_line += spaces;
+				buffer->current_length = spaces;
+			} else {
+				/* one an empty line this is the same as \n */
+				rc = sclp_initialize_mto(buffer,
+							 buffer->columns);
+				if (rc)
+					return i_msg;
+				sclp_finalize_mto(buffer);
+			}
+			break;
+		case '\b':	/* backspace  */
+			/* "go to (actual column - 1, actual line)" */
+			/* decrement counter indicating position, */
+			/* do not remove last character */
+			if (buffer->current_line != NULL &&
+			    buffer->current_length > 0) {
+				buffer->current_length--;
+				buffer->current_line--;
+			}
+			break;
+		case 0x00:	/* end of string  */
+			/* transfer current line to SCCB */
+			if (buffer->current_line != NULL)
+				sclp_finalize_mto(buffer);
+			/* skip the rest of the message including the 0 byte */
+			i_msg = count - 1;
+			break;
+		default:	/* no escape character	*/
+			/* do not output unprintable characters */
+			if (!isprint(msg[i_msg]))
+				break;
+			/* check if new mto needs to be created */
+			if (buffer->current_line == NULL) {
+				rc = sclp_initialize_mto(buffer,
+							 buffer->columns);
+				if (rc)
+					return i_msg;
+			}
+			*buffer->current_line++ = sclp_ascebc(msg[i_msg]);
+			buffer->current_length++;
+			break;
+		}
+		/* check if current mto is full */
+		if (buffer->current_line != NULL &&
+		    buffer->current_length >= buffer->columns)
+			sclp_finalize_mto(buffer);
+	}
+
+	/* return number of processed characters */
+	return i_msg;
+}
+
+/*
+ * Return the number of free bytes in the sccb
+ */
+int
+sclp_buffer_space(struct sclp_buffer *buffer)
+{
+	int count;
+
+	count = MAX_SCCB_ROOM - buffer->sccb->header.length;
+	if (buffer->current_line != NULL)
+		count -= sizeof(struct mto) + buffer->current_length;
+	return count;
+}
+
+/*
+ * Return number of characters in buffer
+ */
+int
+sclp_chars_in_buffer(struct sclp_buffer *buffer)
+{
+	int count;
+
+	count = buffer->mto_char_sum;
+	if (buffer->current_line != NULL)
+		count += buffer->current_length;
+	return count;
+}
+
+/*
+ * sets or provides some values that influence the drivers behaviour
+ */
+void
+sclp_set_columns(struct sclp_buffer *buffer, unsigned short columns)
+{
+	buffer->columns = columns;
+	if (buffer->current_line != NULL &&
+	    buffer->current_length > buffer->columns)
+		sclp_finalize_mto(buffer);
+}
+
+void
+sclp_set_htab(struct sclp_buffer *buffer, unsigned short htab)
+{
+	buffer->htab = htab;
+}
+
+/*
+ * called by sclp_console_init and/or sclp_tty_init
+ */
+int
+sclp_rw_init(void)
+{
+	static int init_done = 0;
+	int rc;
+
+	if (init_done)
+		return 0;
+
+	rc = sclp_register(&sclp_rw_event);
+	if (rc == 0)
+		init_done = 1;
+	return rc;
+}
+
+#define SCLP_BUFFER_MAX_RETRY		1
+
+/*
+ * second half of Write Event Data-function that has to be done after
+ * interruption indicating completion of Service Call.
+ */
+static void
+sclp_writedata_callback(struct sclp_req *request, void *data)
+{
+	int rc;
+	struct sclp_buffer *buffer;
+	struct write_sccb *sccb;
+
+	buffer = (struct sclp_buffer *) data;
+	sccb = buffer->sccb;
+
+	if (request->status == SCLP_REQ_FAILED) {
+		if (buffer->callback != NULL)
+			buffer->callback(buffer, -EIO);
+		return;
+	}
+	/* check SCLP response code and choose suitable action	*/
+	switch (sccb->header.response_code) {
+	case 0x0020 :
+		/* Normal completion, buffer processed, message(s) sent */
+		rc = 0;
+		break;
+
+	case 0x0340: /* Contained SCLP equipment check */
+		if (++buffer->retry_count > SCLP_BUFFER_MAX_RETRY) {
+			rc = -EIO;
+			break;
+		}
+		/* remove processed buffers and requeue rest */
+		if (sclp_remove_processed((struct sccb_header *) sccb) > 0) {
+			/* not all buffers were processed */
+			sccb->header.response_code = 0x0000;
+			buffer->request.status = SCLP_REQ_FILLED;
+			rc = sclp_add_request(request);
+			if (rc == 0)
+				return;
+		} else
+			rc = 0;
+		break;
+
+	case 0x0040: /* SCLP equipment check */
+	case 0x05f0: /* Target resource in improper state */
+		if (++buffer->retry_count > SCLP_BUFFER_MAX_RETRY) {
+			rc = -EIO;
+			break;
+		}
+		/* retry request */
+		sccb->header.response_code = 0x0000;
+		buffer->request.status = SCLP_REQ_FILLED;
+		rc = sclp_add_request(request);
+		if (rc == 0)
+			return;
+		break;
+	default:
+		if (sccb->header.response_code == 0x71f0)
+			rc = -ENOMEM;
+		else
+			rc = -EINVAL;
+		break;
+	}
+	if (buffer->callback != NULL)
+		buffer->callback(buffer, rc);
+}
+
+/*
+ * Setup the request structure in the struct sclp_buffer to do SCLP Write
+ * Event Data and pass the request to the core SCLP loop. Return zero on
+ * success, non-zero otherwise.
+ */
+int
+sclp_emit_buffer(struct sclp_buffer *buffer,
+		 void (*callback)(struct sclp_buffer *, int))
+{
+	struct write_sccb *sccb;
+
+	/* add current line if there is one */
+	if (buffer->current_line != NULL)
+		sclp_finalize_mto(buffer);
+
+	/* Are there messages in the output buffer ? */
+	if (buffer->mto_number == 0)
+		return -EIO;
+
+	sccb = buffer->sccb;
+	if (sclp_rw_event.sclp_receive_mask & EVTYP_MSG_MASK)
+		/* Use normal write message */
+		sccb->msg_buf.header.type = EVTYP_MSG;
+	else if (sclp_rw_event.sclp_receive_mask & EVTYP_PMSGCMD_MASK)
+		/* Use write priority message */
+		sccb->msg_buf.header.type = EVTYP_PMSGCMD;
+	else
+		return -ENOSYS;
+	buffer->request.command = SCLP_CMDW_WRITE_EVENT_DATA;
+	buffer->request.status = SCLP_REQ_FILLED;
+	buffer->request.callback = sclp_writedata_callback;
+	buffer->request.callback_data = buffer;
+	buffer->request.sccb = sccb;
+	buffer->callback = callback;
+	return sclp_add_request(&buffer->request);
+}

kernel/drivers/s390/char/sclp_rw.h

@@ -0,0 +1,101 @@
+/*
+ * interface to the SCLP-read/write driver
+ *
+ * Copyright IBM Corporation 1999, 2009
+ *
+ * Author(s): Martin Peschke <mpeschke@de.ibm.com>
+ *	      Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#ifndef __SCLP_RW_H__
+#define __SCLP_RW_H__
+
+#include <linux/list.h>
+
+struct mto {
+	u16 length;
+	u16 type;
+	u16 line_type_flags;
+	u8  alarm_control;
+	u8  _reserved[3];
+} __attribute__((packed));
+
+struct go {
+	u16 length;
+	u16 type;
+	u32 domid;
+	u8  hhmmss_time[8];
+	u8  th_time[3];
+	u8  reserved_0;
+	u8  dddyyyy_date[7];
+	u8  _reserved_1;
+	u16 general_msg_flags;
+	u8  _reserved_2[10];
+	u8  originating_system_name[8];
+	u8  job_guest_name[8];
+} __attribute__((packed));
+
+struct mdb_header {
+	u16 length;
+	u16 type;
+	u32 tag;
+	u32 revision_code;
+} __attribute__((packed));
+
+struct mdb {
+	struct mdb_header header;
+	struct go go;
+} __attribute__((packed));
+
+struct msg_buf {
+	struct evbuf_header header;
+	struct mdb mdb;
+} __attribute__((packed));
+
+struct write_sccb {
+	struct sccb_header header;
+	struct msg_buf msg_buf;
+} __attribute__((packed));
+
+/* The number of empty mto buffers that can be contained in a single sccb. */
+#define NR_EMPTY_MTO_PER_SCCB ((PAGE_SIZE - sizeof(struct sclp_buffer) - \
+			sizeof(struct write_sccb)) / sizeof(struct mto))
+
+/*
+ * data structure for information about list of SCCBs (only for writing),
+ * will be located at the end of a SCCBs page
+ */
+struct sclp_buffer {
+	struct list_head list;		/* list_head for sccb_info chain */
+	struct sclp_req request;
+	struct write_sccb *sccb;
+	char *current_line;
+	int current_length;
+	int retry_count;
+	/* output format settings */
+	unsigned short columns;
+	unsigned short htab;
+	/* statistics about this buffer */
+	unsigned int mto_char_sum;	/* # chars in sccb */
+	unsigned int mto_number;	/* # mtos in sccb */
+	/* Callback that is called after reaching final status. */
+	void (*callback)(struct sclp_buffer *, int);
+};
+
+int sclp_rw_init(void);
+struct sclp_buffer *sclp_make_buffer(void *, unsigned short, unsigned short);
+void *sclp_unmake_buffer(struct sclp_buffer *);
+int sclp_buffer_space(struct sclp_buffer *);
+int sclp_write(struct sclp_buffer *buffer, const unsigned char *, int);
+int sclp_emit_buffer(struct sclp_buffer *,void (*)(struct sclp_buffer *,int));
+void sclp_set_columns(struct sclp_buffer *, unsigned short);
+void sclp_set_htab(struct sclp_buffer *, unsigned short);
+int sclp_chars_in_buffer(struct sclp_buffer *);
+
+#ifdef CONFIG_SCLP_CONSOLE
+void sclp_console_pm_event(enum sclp_pm_event sclp_pm_event);
+#else
+static inline void sclp_console_pm_event(enum sclp_pm_event sclp_pm_event) { }
+#endif
+
+#endif	/* __SCLP_RW_H__ */

kernel/drivers/s390/char/sclp_sdias.c

@@ -0,0 +1,257 @@
+/*
+ * Sclp "store data in absolut storage"
+ *
+ * Copyright IBM Corp. 2003,2007
+ * Author(s): Michael Holzheu
+ */
+
+#define KMSG_COMPONENT "sclp_sdias"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/sched.h>
+#include <asm/sclp.h>
+#include <asm/debug.h>
+#include <asm/ipl.h>
+
+#include "sclp.h"
+#include "sclp_rw.h"
+
+#define TRACE(x...) debug_sprintf_event(sdias_dbf, 1, x)
+
+#define SDIAS_RETRIES 300
+#define SDIAS_SLEEP_TICKS 50
+
+#define EQ_STORE_DATA	0x0
+#define EQ_SIZE		0x1
+#define DI_FCP_DUMP	0x0
+#define ASA_SIZE_32	0x0
+#define ASA_SIZE_64	0x1
+#define EVSTATE_ALL_STORED	0x0
+#define EVSTATE_NO_DATA		0x3
+#define EVSTATE_PART_STORED	0x10
+
+static struct debug_info *sdias_dbf;
+
+static struct sclp_register sclp_sdias_register = {
+	.send_mask = EVTYP_SDIAS_MASK,
+};
+
+struct sdias_evbuf {
+	struct	evbuf_header hdr;
+	u8	event_qual;
+	u8	data_id;
+	u64	reserved2;
+	u32	event_id;
+	u16	reserved3;
+	u8	asa_size;
+	u8	event_status;
+	u32	reserved4;
+	u32	blk_cnt;
+	u64	asa;
+	u32	reserved5;
+	u32	fbn;
+	u32	reserved6;
+	u32	lbn;
+	u16	reserved7;
+	u16	dbs;
+} __attribute__((packed));
+
+struct sdias_sccb {
+	struct sccb_header  hdr;
+	struct sdias_evbuf  evbuf;
+} __attribute__((packed));
+
+static struct sdias_sccb sccb __attribute__((aligned(4096)));
+
+static int sclp_req_done;
+static wait_queue_head_t sdias_wq;
+static DEFINE_MUTEX(sdias_mutex);
+
+static void sdias_callback(struct sclp_req *request, void *data)
+{
+	struct sdias_sccb *cbsccb;
+
+	cbsccb = (struct sdias_sccb *) request->sccb;
+	sclp_req_done = 1;
+	wake_up(&sdias_wq); /* Inform caller, that request is complete */
+	TRACE("callback done\n");
+}
+
+static int sdias_sclp_send(struct sclp_req *req)
+{
+	int retries;
+	int rc;
+
+	for (retries = SDIAS_RETRIES; retries; retries--) {
+		sclp_req_done = 0;
+		TRACE("add request\n");
+		rc = sclp_add_request(req);
+		if (rc) {
+			/* not initiated, wait some time and retry */
+			set_current_state(TASK_INTERRUPTIBLE);
+			TRACE("add request failed: rc = %i\n",rc);
+			schedule_timeout(SDIAS_SLEEP_TICKS);
+			continue;
+		}
+		/* initiated, wait for completion of service call */
+		wait_event(sdias_wq, (sclp_req_done == 1));
+		if (req->status == SCLP_REQ_FAILED) {
+			TRACE("sclp request failed\n");
+			rc = -EIO;
+			continue;
+		}
+		TRACE("request done\n");
+		break;
+	}
+	return rc;
+}
+
+/*
+ * Get number of blocks (4K) available in the HSA
+ */
+int sclp_sdias_blk_count(void)
+{
+	struct sclp_req request;
+	int rc;
+
+	mutex_lock(&sdias_mutex);
+
+	memset(&sccb, 0, sizeof(sccb));
+	memset(&request, 0, sizeof(request));
+
+	sccb.hdr.length = sizeof(sccb);
+	sccb.evbuf.hdr.length = sizeof(struct sdias_evbuf);
+	sccb.evbuf.hdr.type = EVTYP_SDIAS;
+	sccb.evbuf.event_qual = EQ_SIZE;
+	sccb.evbuf.data_id = DI_FCP_DUMP;
+	sccb.evbuf.event_id = 4712;
+	sccb.evbuf.dbs = 1;
+
+	request.sccb = &sccb;
+	request.command = SCLP_CMDW_WRITE_EVENT_DATA;
+	request.status = SCLP_REQ_FILLED;
+	request.callback = sdias_callback;
+
+	rc = sdias_sclp_send(&request);
+	if (rc) {
+		pr_err("sclp_send failed for get_nr_blocks\n");
+		goto out;
+	}
+	if (sccb.hdr.response_code != 0x0020) {
+		TRACE("send failed: %x\n", sccb.hdr.response_code);
+		rc = -EIO;
+		goto out;
+	}
+
+	switch (sccb.evbuf.event_status) {
+		case 0:
+			rc = sccb.evbuf.blk_cnt;
+			break;
+		default:
+			pr_err("SCLP error: %x\n",
+			       sccb.evbuf.event_status);
+			rc = -EIO;
+			goto out;
+	}
+	TRACE("%i blocks\n", rc);
+out:
+	mutex_unlock(&sdias_mutex);
+	return rc;
+}
+
+/*
+ * Copy from HSA to absolute storage (not reentrant):
+ *
+ * @dest     : Address of buffer where data should be copied
+ * @start_blk: Start Block (beginning with 1)
+ * @nr_blks  : Number of 4K blocks to copy
+ *
+ * Return Value: 0 : Requested 'number' of blocks of data copied
+ *		 <0: ERROR - negative event status
+ */
+int sclp_sdias_copy(void *dest, int start_blk, int nr_blks)
+{
+	struct sclp_req request;
+	int rc;
+
+	mutex_lock(&sdias_mutex);
+
+	memset(&sccb, 0, sizeof(sccb));
+	memset(&request, 0, sizeof(request));
+
+	sccb.hdr.length = sizeof(sccb);
+	sccb.evbuf.hdr.length = sizeof(struct sdias_evbuf);
+	sccb.evbuf.hdr.type = EVTYP_SDIAS;
+	sccb.evbuf.hdr.flags = 0;
+	sccb.evbuf.event_qual = EQ_STORE_DATA;
+	sccb.evbuf.data_id = DI_FCP_DUMP;
+	sccb.evbuf.event_id = 4712;
+#ifdef __s390x__
+	sccb.evbuf.asa_size = ASA_SIZE_64;
+#else
+	sccb.evbuf.asa_size = ASA_SIZE_32;
+#endif
+	sccb.evbuf.event_status = 0;
+	sccb.evbuf.blk_cnt = nr_blks;
+	sccb.evbuf.asa = (unsigned long)dest;
+	sccb.evbuf.fbn = start_blk;
+	sccb.evbuf.lbn = 0;
+	sccb.evbuf.dbs = 1;
+
+	request.sccb	 = &sccb;
+	request.command  = SCLP_CMDW_WRITE_EVENT_DATA;
+	request.status	 = SCLP_REQ_FILLED;
+	request.callback = sdias_callback;
+
+	rc = sdias_sclp_send(&request);
+	if (rc) {
+		pr_err("sclp_send failed: %x\n", rc);
+		goto out;
+	}
+	if (sccb.hdr.response_code != 0x0020) {
+		TRACE("copy failed: %x\n", sccb.hdr.response_code);
+		rc = -EIO;
+		goto out;
+	}
+
+	switch (sccb.evbuf.event_status) {
+		case EVSTATE_ALL_STORED:
+			TRACE("all stored\n");
+		case EVSTATE_PART_STORED:
+			TRACE("part stored: %i\n", sccb.evbuf.blk_cnt);
+			break;
+		case EVSTATE_NO_DATA:
+			TRACE("no data\n");
+		default:
+			pr_err("Error from SCLP while copying hsa. "
+			       "Event status = %x\n",
+			       sccb.evbuf.event_status);
+			rc = -EIO;
+	}
+out:
+	mutex_unlock(&sdias_mutex);
+	return rc;
+}
+
+int __init sclp_sdias_init(void)
+{
+	int rc;
+
+	if (ipl_info.type != IPL_TYPE_FCP_DUMP)
+		return 0;
+	sdias_dbf = debug_register("dump_sdias", 4, 1, 4 * sizeof(long));
+	debug_register_view(sdias_dbf, &debug_sprintf_view);
+	debug_set_level(sdias_dbf, 6);
+	rc = sclp_register(&sclp_sdias_register);
+	if (rc)
+		return rc;
+	init_waitqueue_head(&sdias_wq);
+	TRACE("init done\n");
+	return 0;
+}
+
+void __exit sclp_sdias_exit(void)
+{
+	debug_unregister(sdias_dbf);
+	sclp_unregister(&sclp_sdias_register);
+}

kernel/drivers/s390/char/sclp_tty.c

@@ -0,0 +1,621 @@
+/*
+ *  drivers/s390/char/sclp_tty.c
+ *    SCLP line mode terminal driver.
+ *
+ *  S390 version
+ *    Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Author(s): Martin Peschke <mpeschke@de.ibm.com>
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#include <linux/module.h>
+#include <linux/kmod.h>
+#include <linux/tty.h>
+#include <linux/tty_driver.h>
+#include <linux/tty_flip.h>
+#include <linux/slab.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <asm/uaccess.h>
+
+#include "ctrlchar.h"
+#include "sclp.h"
+#include "sclp_rw.h"
+#include "sclp_tty.h"
+
+/*
+ * size of a buffer that collects single characters coming in
+ * via sclp_tty_put_char()
+ */
+#define SCLP_TTY_BUF_SIZE 512
+
+/*
+ * There is exactly one SCLP terminal, so we can keep things simple
+ * and allocate all variables statically.
+ */
+
+/* Lock to guard over changes to global variables. */
+static spinlock_t sclp_tty_lock;
+/* List of free pages that can be used for console output buffering. */
+static struct list_head sclp_tty_pages;
+/* List of full struct sclp_buffer structures ready for output. */
+static struct list_head sclp_tty_outqueue;
+/* Counter how many buffers are emitted. */
+static int sclp_tty_buffer_count;
+/* Pointer to current console buffer. */
+static struct sclp_buffer *sclp_ttybuf;
+/* Timer for delayed output of console messages. */
+static struct timer_list sclp_tty_timer;
+
+static struct tty_struct *sclp_tty;
+static unsigned char sclp_tty_chars[SCLP_TTY_BUF_SIZE];
+static unsigned short int sclp_tty_chars_count;
+
+struct tty_driver *sclp_tty_driver;
+
+static int sclp_tty_tolower;
+static int sclp_tty_columns = 80;
+
+#define SPACES_PER_TAB 8
+#define CASE_DELIMITER 0x6c /* to separate upper and lower case (% in EBCDIC) */
+
+/* This routine is called whenever we try to open a SCLP terminal. */
+static int
+sclp_tty_open(struct tty_struct *tty, struct file *filp)
+{
+	sclp_tty = tty;
+	tty->driver_data = NULL;
+	tty->low_latency = 0;
+	return 0;
+}
+
+/* This routine is called when the SCLP terminal is closed. */
+static void
+sclp_tty_close(struct tty_struct *tty, struct file *filp)
+{
+	if (tty->count > 1)
+		return;
+	sclp_tty = NULL;
+}
+
+/*
+ * This routine returns the numbers of characters the tty driver
+ * will accept for queuing to be written.  This number is subject
+ * to change as output buffers get emptied, or if the output flow
+ * control is acted. This is not an exact number because not every
+ * character needs the same space in the sccb. The worst case is
+ * a string of newlines. Every newlines creates a new mto which
+ * needs 8 bytes.
+ */
+static int
+sclp_tty_write_room (struct tty_struct *tty)
+{
+	unsigned long flags;
+	struct list_head *l;
+	int count;
+
+	spin_lock_irqsave(&sclp_tty_lock, flags);
+	count = 0;
+	if (sclp_ttybuf != NULL)
+		count = sclp_buffer_space(sclp_ttybuf) / sizeof(struct mto);
+	list_for_each(l, &sclp_tty_pages)
+		count += NR_EMPTY_MTO_PER_SCCB;
+	spin_unlock_irqrestore(&sclp_tty_lock, flags);
+	return count;
+}
+
+static void
+sclp_ttybuf_callback(struct sclp_buffer *buffer, int rc)
+{
+	unsigned long flags;
+	void *page;
+
+	do {
+		page = sclp_unmake_buffer(buffer);
+		spin_lock_irqsave(&sclp_tty_lock, flags);
+		/* Remove buffer from outqueue */
+		list_del(&buffer->list);
+		sclp_tty_buffer_count--;
+		list_add_tail((struct list_head *) page, &sclp_tty_pages);
+		/* Check if there is a pending buffer on the out queue. */
+		buffer = NULL;
+		if (!list_empty(&sclp_tty_outqueue))
+			buffer = list_entry(sclp_tty_outqueue.next,
+					    struct sclp_buffer, list);
+		spin_unlock_irqrestore(&sclp_tty_lock, flags);
+	} while (buffer && sclp_emit_buffer(buffer, sclp_ttybuf_callback));
+	/* check if the tty needs a wake up call */
+	if (sclp_tty != NULL) {
+		tty_wakeup(sclp_tty);
+	}
+}
+
+static inline void
+__sclp_ttybuf_emit(struct sclp_buffer *buffer)
+{
+	unsigned long flags;
+	int count;
+	int rc;
+
+	spin_lock_irqsave(&sclp_tty_lock, flags);
+	list_add_tail(&buffer->list, &sclp_tty_outqueue);
+	count = sclp_tty_buffer_count++;
+	spin_unlock_irqrestore(&sclp_tty_lock, flags);
+	if (count)
+		return;
+	rc = sclp_emit_buffer(buffer, sclp_ttybuf_callback);
+	if (rc)
+		sclp_ttybuf_callback(buffer, rc);
+}
+
+/*
+ * When this routine is called from the timer then we flush the
+ * temporary write buffer.
+ */
+static void
+sclp_tty_timeout(unsigned long data)
+{
+	unsigned long flags;
+	struct sclp_buffer *buf;
+
+	spin_lock_irqsave(&sclp_tty_lock, flags);
+	buf = sclp_ttybuf;
+	sclp_ttybuf = NULL;
+	spin_unlock_irqrestore(&sclp_tty_lock, flags);
+
+	if (buf != NULL) {
+		__sclp_ttybuf_emit(buf);
+	}
+}
+
+/*
+ * Write a string to the sclp tty.
+ */
+static int sclp_tty_write_string(const unsigned char *str, int count, int may_fail)
+{
+	unsigned long flags;
+	void *page;
+	int written;
+	int overall_written;
+	struct sclp_buffer *buf;
+
+	if (count <= 0)
+		return 0;
+	overall_written = 0;
+	spin_lock_irqsave(&sclp_tty_lock, flags);
+	do {
+		/* Create a sclp output buffer if none exists yet */
+		if (sclp_ttybuf == NULL) {
+			while (list_empty(&sclp_tty_pages)) {
+				spin_unlock_irqrestore(&sclp_tty_lock, flags);
+				if (may_fail)
+					goto out;
+				else
+					sclp_sync_wait();
+				spin_lock_irqsave(&sclp_tty_lock, flags);
+			}
+			page = sclp_tty_pages.next;
+			list_del((struct list_head *) page);
+			sclp_ttybuf = sclp_make_buffer(page, sclp_tty_columns,
+						       SPACES_PER_TAB);
+		}
+		/* try to write the string to the current output buffer */
+		written = sclp_write(sclp_ttybuf, str, count);
+		overall_written += written;
+		if (written == count)
+			break;
+		/*
+		 * Not all characters could be written to the current
+		 * output buffer. Emit the buffer, create a new buffer
+		 * and then output the rest of the string.
+		 */
+		buf = sclp_ttybuf;
+		sclp_ttybuf = NULL;
+		spin_unlock_irqrestore(&sclp_tty_lock, flags);
+		__sclp_ttybuf_emit(buf);
+		spin_lock_irqsave(&sclp_tty_lock, flags);
+		str += written;
+		count -= written;
+	} while (count > 0);
+	/* Setup timer to output current console buffer after 1/10 second */
+	if (sclp_ttybuf && sclp_chars_in_buffer(sclp_ttybuf) &&
+	    !timer_pending(&sclp_tty_timer)) {
+		init_timer(&sclp_tty_timer);
+		sclp_tty_timer.function = sclp_tty_timeout;
+		sclp_tty_timer.data = 0UL;
+		sclp_tty_timer.expires = jiffies + HZ/10;
+		add_timer(&sclp_tty_timer);
+	}
+	spin_unlock_irqrestore(&sclp_tty_lock, flags);
+out:
+	return overall_written;
+}
+
+/*
+ * This routine is called by the kernel to write a series of characters to the
+ * tty device. The characters may come from user space or kernel space. This
+ * routine will return the number of characters actually accepted for writing.
+ */
+static int
+sclp_tty_write(struct tty_struct *tty, const unsigned char *buf, int count)
+{
+	if (sclp_tty_chars_count > 0) {
+		sclp_tty_write_string(sclp_tty_chars, sclp_tty_chars_count, 0);
+		sclp_tty_chars_count = 0;
+	}
+	return sclp_tty_write_string(buf, count, 1);
+}
+
+/*
+ * This routine is called by the kernel to write a single character to the tty
+ * device. If the kernel uses this routine, it must call the flush_chars()
+ * routine (if defined) when it is done stuffing characters into the driver.
+ *
+ * Characters provided to sclp_tty_put_char() are buffered by the SCLP driver.
+ * If the given character is a '\n' the contents of the SCLP write buffer
+ * - including previous characters from sclp_tty_put_char() and strings from
+ * sclp_write() without final '\n' - will be written.
+ */
+static int
+sclp_tty_put_char(struct tty_struct *tty, unsigned char ch)
+{
+	sclp_tty_chars[sclp_tty_chars_count++] = ch;
+	if (ch == '\n' || sclp_tty_chars_count >= SCLP_TTY_BUF_SIZE) {
+		sclp_tty_write_string(sclp_tty_chars, sclp_tty_chars_count, 0);
+		sclp_tty_chars_count = 0;
+	}
+	return 1;
+}
+
+/*
+ * This routine is called by the kernel after it has written a series of
+ * characters to the tty device using put_char().
+ */
+static void
+sclp_tty_flush_chars(struct tty_struct *tty)
+{
+	if (sclp_tty_chars_count > 0) {
+		sclp_tty_write_string(sclp_tty_chars, sclp_tty_chars_count, 0);
+		sclp_tty_chars_count = 0;
+	}
+}
+
+/*
+ * This routine returns the number of characters in the write buffer of the
+ * SCLP driver. The provided number includes all characters that are stored
+ * in the SCCB (will be written next time the SCLP is not busy) as well as
+ * characters in the write buffer (will not be written as long as there is a
+ * final line feed missing).
+ */
+static int
+sclp_tty_chars_in_buffer(struct tty_struct *tty)
+{
+	unsigned long flags;
+	struct list_head *l;
+	struct sclp_buffer *t;
+	int count;
+
+	spin_lock_irqsave(&sclp_tty_lock, flags);
+	count = 0;
+	if (sclp_ttybuf != NULL)
+		count = sclp_chars_in_buffer(sclp_ttybuf);
+	list_for_each(l, &sclp_tty_outqueue) {
+		t = list_entry(l, struct sclp_buffer, list);
+		count += sclp_chars_in_buffer(t);
+	}
+	spin_unlock_irqrestore(&sclp_tty_lock, flags);
+	return count;
+}
+
+/*
+ * removes all content from buffers of low level driver
+ */
+static void
+sclp_tty_flush_buffer(struct tty_struct *tty)
+{
+	if (sclp_tty_chars_count > 0) {
+		sclp_tty_write_string(sclp_tty_chars, sclp_tty_chars_count, 0);
+		sclp_tty_chars_count = 0;
+	}
+}
+
+/*
+ * push input to tty
+ */
+static void
+sclp_tty_input(unsigned char* buf, unsigned int count)
+{
+	unsigned int cchar;
+
+	/*
+	 * If this tty driver is currently closed
+	 * then throw the received input away.
+	 */
+	if (sclp_tty == NULL)
+		return;
+	cchar = ctrlchar_handle(buf, count, sclp_tty);
+	switch (cchar & CTRLCHAR_MASK) {
+	case CTRLCHAR_SYSRQ:
+		break;
+	case CTRLCHAR_CTRL:
+		tty_insert_flip_char(sclp_tty, cchar, TTY_NORMAL);
+		tty_flip_buffer_push(sclp_tty);
+		break;
+	case CTRLCHAR_NONE:
+		/* send (normal) input to line discipline */
+		if (count < 2 ||
+		    (strncmp((const char *) buf + count - 2, "^n", 2) &&
+		     strncmp((const char *) buf + count - 2, "\252n", 2))) {
+			/* add the auto \n */
+			tty_insert_flip_string(sclp_tty, buf, count);
+			tty_insert_flip_char(sclp_tty, '\n', TTY_NORMAL);
+		} else
+			tty_insert_flip_string(sclp_tty, buf, count - 2);
+		tty_flip_buffer_push(sclp_tty);
+		break;
+	}
+}
+
+/*
+ * get a EBCDIC string in upper/lower case,
+ * find out characters in lower/upper case separated by a special character,
+ * modifiy original string,
+ * returns length of resulting string
+ */
+static int sclp_switch_cases(unsigned char *buf, int count)
+{
+	unsigned char *ip, *op;
+	int toggle;
+
+	/* initially changing case is off */
+	toggle = 0;
+	ip = op = buf;
+	while (count-- > 0) {
+		/* compare with special character */
+		if (*ip == CASE_DELIMITER) {
+			/* followed by another special character? */
+			if (count && ip[1] == CASE_DELIMITER) {
+				/*
+				 * ... then put a single copy of the special
+				 * character to the output string
+				 */
+				*op++ = *ip++;
+				count--;
+			} else
+				/*
+				 * ... special character follower by a normal
+				 * character toggles the case change behaviour
+				 */
+				toggle = ~toggle;
+			/* skip special character */
+			ip++;
+		} else
+			/* not the special character */
+			if (toggle)
+				/* but case switching is on */
+				if (sclp_tty_tolower)
+					/* switch to uppercase */
+					*op++ = _ebc_toupper[(int) *ip++];
+				else
+					/* switch to lowercase */
+					*op++ = _ebc_tolower[(int) *ip++];
+			else
+				/* no case switching, copy the character */
+				*op++ = *ip++;
+	}
+	/* return length of reformatted string. */
+	return op - buf;
+}
+
+static void
+sclp_get_input(unsigned char *start, unsigned char *end)
+{
+	int count;
+
+	count = end - start;
+	if (sclp_tty_tolower)
+		EBC_TOLOWER(start, count);
+	count = sclp_switch_cases(start, count);
+	/* convert EBCDIC to ASCII (modify original input in SCCB) */
+	sclp_ebcasc_str(start, count);
+
+	/* transfer input to high level driver */
+	sclp_tty_input(start, count);
+}
+
+static inline struct gds_vector *
+find_gds_vector(struct gds_vector *start, struct gds_vector *end, u16 id)
+{
+	struct gds_vector *vec;
+
+	for (vec = start; vec < end; vec = (void *) vec + vec->length)
+		if (vec->gds_id == id)
+			return vec;
+	return NULL;
+}
+
+static inline struct gds_subvector *
+find_gds_subvector(struct gds_subvector *start,
+		   struct gds_subvector *end, u8 key)
+{
+	struct gds_subvector *subvec;
+
+	for (subvec = start; subvec < end;
+	     subvec = (void *) subvec + subvec->length)
+		if (subvec->key == key)
+			return subvec;
+	return NULL;
+}
+
+static inline void
+sclp_eval_selfdeftextmsg(struct gds_subvector *start,
+			 struct gds_subvector *end)
+{
+	struct gds_subvector *subvec;
+
+	subvec = start;
+	while (subvec < end) {
+		subvec = find_gds_subvector(subvec, end, 0x30);
+		if (!subvec)
+			break;
+		sclp_get_input((unsigned char *)(subvec + 1),
+			       (unsigned char *) subvec + subvec->length);
+		subvec = (void *) subvec + subvec->length;
+	}
+}
+
+static inline void
+sclp_eval_textcmd(struct gds_subvector *start,
+		  struct gds_subvector *end)
+{
+	struct gds_subvector *subvec;
+
+	subvec = start;
+	while (subvec < end) {
+		subvec = find_gds_subvector(subvec, end,
+					    GDS_KEY_SELFDEFTEXTMSG);
+		if (!subvec)
+			break;
+		sclp_eval_selfdeftextmsg((struct gds_subvector *)(subvec + 1),
+					 (void *)subvec + subvec->length);
+		subvec = (void *) subvec + subvec->length;
+	}
+}
+
+static inline void
+sclp_eval_cpmsu(struct gds_vector *start, struct gds_vector *end)
+{
+	struct gds_vector *vec;
+
+	vec = start;
+	while (vec < end) {
+		vec = find_gds_vector(vec, end, GDS_ID_TEXTCMD);
+		if (!vec)
+			break;
+		sclp_eval_textcmd((struct gds_subvector *)(vec + 1),
+				  (void *) vec + vec->length);
+		vec = (void *) vec + vec->length;
+	}
+}
+
+
+static inline void
+sclp_eval_mdsmu(struct gds_vector *start, void *end)
+{
+	struct gds_vector *vec;
+
+	vec = find_gds_vector(start, end, GDS_ID_CPMSU);
+	if (vec)
+		sclp_eval_cpmsu(vec + 1, (void *) vec + vec->length);
+}
+
+static void
+sclp_tty_receiver(struct evbuf_header *evbuf)
+{
+	struct gds_vector *start, *end, *vec;
+
+	start = (struct gds_vector *)(evbuf + 1);
+	end = (void *) evbuf + evbuf->length;
+	vec = find_gds_vector(start, end, GDS_ID_MDSMU);
+	if (vec)
+		sclp_eval_mdsmu(vec + 1, (void *) vec + vec->length);
+}
+
+static void
+sclp_tty_state_change(struct sclp_register *reg)
+{
+}
+
+static struct sclp_register sclp_input_event =
+{
+	.receive_mask = EVTYP_OPCMD_MASK | EVTYP_PMSGCMD_MASK,
+	.state_change_fn = sclp_tty_state_change,
+	.receiver_fn = sclp_tty_receiver
+};
+
+static const struct tty_operations sclp_ops = {
+	.open = sclp_tty_open,
+	.close = sclp_tty_close,
+	.write = sclp_tty_write,
+	.put_char = sclp_tty_put_char,
+	.flush_chars = sclp_tty_flush_chars,
+	.write_room = sclp_tty_write_room,
+	.chars_in_buffer = sclp_tty_chars_in_buffer,
+	.flush_buffer = sclp_tty_flush_buffer,
+};
+
+static int __init
+sclp_tty_init(void)
+{
+	struct tty_driver *driver;
+	void *page;
+	int i;
+	int rc;
+
+	if (!CONSOLE_IS_SCLP)
+		return 0;
+	driver = alloc_tty_driver(1);
+	if (!driver)
+		return -ENOMEM;
+
+	rc = sclp_rw_init();
+	if (rc) {
+		put_tty_driver(driver);
+		return rc;
+	}
+	/* Allocate pages for output buffering */
+	INIT_LIST_HEAD(&sclp_tty_pages);
+	for (i = 0; i < MAX_KMEM_PAGES; i++) {
+		page = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		if (page == NULL) {
+			put_tty_driver(driver);
+			return -ENOMEM;
+		}
+		list_add_tail((struct list_head *) page, &sclp_tty_pages);
+	}
+	INIT_LIST_HEAD(&sclp_tty_outqueue);
+	spin_lock_init(&sclp_tty_lock);
+	init_timer(&sclp_tty_timer);
+	sclp_ttybuf = NULL;
+	sclp_tty_buffer_count = 0;
+	if (MACHINE_IS_VM) {
+		/*
+		 * save 4 characters for the CPU number
+		 * written at start of each line by VM/CP
+		 */
+		sclp_tty_columns = 76;
+		/* case input lines to lowercase */
+		sclp_tty_tolower = 1;
+	}
+	sclp_tty_chars_count = 0;
+	sclp_tty = NULL;
+
+	rc = sclp_register(&sclp_input_event);
+	if (rc) {
+		put_tty_driver(driver);
+		return rc;
+	}
+
+	driver->owner = THIS_MODULE;
+	driver->driver_name = "sclp_line";
+	driver->name = "sclp_line";
+	driver->major = TTY_MAJOR;
+	driver->minor_start = 64;
+	driver->type = TTY_DRIVER_TYPE_SYSTEM;
+	driver->subtype = SYSTEM_TYPE_TTY;
+	driver->init_termios = tty_std_termios;
+	driver->init_termios.c_iflag = IGNBRK | IGNPAR;
+	driver->init_termios.c_oflag = ONLCR | XTABS;
+	driver->init_termios.c_lflag = ISIG | ECHO;
+	driver->flags = TTY_DRIVER_REAL_RAW;
+	tty_set_operations(driver, &sclp_ops);
+	rc = tty_register_driver(driver);
+	if (rc) {
+		put_tty_driver(driver);
+		return rc;
+	}
+	sclp_tty_driver = driver;
+	return 0;
+}
+module_init(sclp_tty_init);

kernel/drivers/s390/char/sclp_tty.h

@@ -0,0 +1,18 @@
+/*
+ *  drivers/s390/char/sclp_tty.h
+ *    interface to the SCLP-read/write driver
+ *
+ *  S390 version
+ *    Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Author(s): Martin Peschke <mpeschke@de.ibm.com>
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#ifndef __SCLP_TTY_H__
+#define __SCLP_TTY_H__
+
+#include <linux/tty_driver.h>
+
+extern struct tty_driver *sclp_tty_driver;
+
+#endif	/* __SCLP_TTY_H__ */

kernel/drivers/s390/char/sclp_vt220.c

@@ -0,0 +1,826 @@
+/*
+ * SCLP VT220 terminal driver.
+ *
+ * Copyright IBM Corp. 2003, 2009
+ *
+ * Author(s): Peter Oberparleiter <Peter.Oberparleiter@de.ibm.com>
+ */
+
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/list.h>
+#include <linux/wait.h>
+#include <linux/timer.h>
+#include <linux/kernel.h>
+#include <linux/tty.h>
+#include <linux/tty_driver.h>
+#include <linux/tty_flip.h>
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/major.h>
+#include <linux/console.h>
+#include <linux/kdev_t.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/reboot.h>
+
+#include <asm/uaccess.h>
+#include "sclp.h"
+
+#define SCLP_VT220_MAJOR		TTY_MAJOR
+#define SCLP_VT220_MINOR		65
+#define SCLP_VT220_DRIVER_NAME		"sclp_vt220"
+#define SCLP_VT220_DEVICE_NAME		"ttysclp"
+#define SCLP_VT220_CONSOLE_NAME		"ttyS"
+#define SCLP_VT220_CONSOLE_INDEX	1	/* console=ttyS1 */
+#define SCLP_VT220_BUF_SIZE		80
+
+/* Representation of a single write request */
+struct sclp_vt220_request {
+	struct list_head list;
+	struct sclp_req sclp_req;
+	int retry_count;
+};
+
+/* VT220 SCCB */
+struct sclp_vt220_sccb {
+	struct sccb_header header;
+	struct evbuf_header evbuf;
+};
+
+#define SCLP_VT220_MAX_CHARS_PER_BUFFER	(PAGE_SIZE - \
+					 sizeof(struct sclp_vt220_request) - \
+					 sizeof(struct sclp_vt220_sccb))
+
+/* Structures and data needed to register tty driver */
+static struct tty_driver *sclp_vt220_driver;
+
+/* The tty_struct that the kernel associated with us */
+static struct tty_struct *sclp_vt220_tty;
+
+/* Lock to protect internal data from concurrent access */
+static spinlock_t sclp_vt220_lock;
+
+/* List of empty pages to be used as write request buffers */
+static struct list_head sclp_vt220_empty;
+
+/* List of pending requests */
+static struct list_head sclp_vt220_outqueue;
+
+/* Suspend mode flag */
+static int sclp_vt220_suspended;
+
+/* Flag that output queue is currently running */
+static int sclp_vt220_queue_running;
+
+/* Timer used for delaying write requests to merge subsequent messages into
+ * a single buffer */
+static struct timer_list sclp_vt220_timer;
+
+/* Pointer to current request buffer which has been partially filled but not
+ * yet sent */
+static struct sclp_vt220_request *sclp_vt220_current_request;
+
+/* Number of characters in current request buffer */
+static int sclp_vt220_buffered_chars;
+
+/* Counter controlling core driver initialization. */
+static int __initdata sclp_vt220_init_count;
+
+/* Flag indicating that sclp_vt220_current_request should really
+ * have been already queued but wasn't because the SCLP was processing
+ * another buffer */
+static int sclp_vt220_flush_later;
+
+static void sclp_vt220_receiver_fn(struct evbuf_header *evbuf);
+static void sclp_vt220_pm_event_fn(struct sclp_register *reg,
+				   enum sclp_pm_event sclp_pm_event);
+static int __sclp_vt220_emit(struct sclp_vt220_request *request);
+static void sclp_vt220_emit_current(void);
+
+/* Registration structure for our interest in SCLP event buffers */
+static struct sclp_register sclp_vt220_register = {
+	.send_mask		= EVTYP_VT220MSG_MASK,
+	.receive_mask		= EVTYP_VT220MSG_MASK,
+	.state_change_fn	= NULL,
+	.receiver_fn		= sclp_vt220_receiver_fn,
+	.pm_event_fn		= sclp_vt220_pm_event_fn,
+};
+
+
+/*
+ * Put provided request buffer back into queue and check emit pending
+ * buffers if necessary.
+ */
+static void
+sclp_vt220_process_queue(struct sclp_vt220_request *request)
+{
+	unsigned long flags;
+	void *page;
+
+	do {
+		/* Put buffer back to list of empty buffers */
+		page = request->sclp_req.sccb;
+		spin_lock_irqsave(&sclp_vt220_lock, flags);
+		/* Move request from outqueue to empty queue */
+		list_del(&request->list);
+		list_add_tail((struct list_head *) page, &sclp_vt220_empty);
+		/* Check if there is a pending buffer on the out queue. */
+		request = NULL;
+		if (!list_empty(&sclp_vt220_outqueue))
+			request = list_entry(sclp_vt220_outqueue.next,
+					     struct sclp_vt220_request, list);
+		if (!request || sclp_vt220_suspended) {
+			sclp_vt220_queue_running = 0;
+			spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+			break;
+		}
+		spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+	} while (__sclp_vt220_emit(request));
+	if (request == NULL && sclp_vt220_flush_later)
+		sclp_vt220_emit_current();
+	/* Check if the tty needs a wake up call */
+	if (sclp_vt220_tty != NULL) {
+		tty_wakeup(sclp_vt220_tty);
+	}
+}
+
+#define SCLP_BUFFER_MAX_RETRY		1
+
+/*
+ * Callback through which the result of a write request is reported by the
+ * SCLP.
+ */
+static void
+sclp_vt220_callback(struct sclp_req *request, void *data)
+{
+	struct sclp_vt220_request *vt220_request;
+	struct sclp_vt220_sccb *sccb;
+
+	vt220_request = (struct sclp_vt220_request *) data;
+	if (request->status == SCLP_REQ_FAILED) {
+		sclp_vt220_process_queue(vt220_request);
+		return;
+	}
+	sccb = (struct sclp_vt220_sccb *) vt220_request->sclp_req.sccb;
+
+	/* Check SCLP response code and choose suitable action	*/
+	switch (sccb->header.response_code) {
+	case 0x0020 :
+		break;
+
+	case 0x05f0: /* Target resource in improper state */
+		break;
+
+	case 0x0340: /* Contained SCLP equipment check */
+		if (++vt220_request->retry_count > SCLP_BUFFER_MAX_RETRY)
+			break;
+		/* Remove processed buffers and requeue rest */
+		if (sclp_remove_processed((struct sccb_header *) sccb) > 0) {
+			/* Not all buffers were processed */
+			sccb->header.response_code = 0x0000;
+			vt220_request->sclp_req.status = SCLP_REQ_FILLED;
+			if (sclp_add_request(request) == 0)
+				return;
+		}
+		break;
+
+	case 0x0040: /* SCLP equipment check */
+		if (++vt220_request->retry_count > SCLP_BUFFER_MAX_RETRY)
+			break;
+		sccb->header.response_code = 0x0000;
+		vt220_request->sclp_req.status = SCLP_REQ_FILLED;
+		if (sclp_add_request(request) == 0)
+			return;
+		break;
+
+	default:
+		break;
+	}
+	sclp_vt220_process_queue(vt220_request);
+}
+
+/*
+ * Emit vt220 request buffer to SCLP. Return zero on success, non-zero
+ * otherwise.
+ */
+static int
+__sclp_vt220_emit(struct sclp_vt220_request *request)
+{
+	if (!(sclp_vt220_register.sclp_receive_mask & EVTYP_VT220MSG_MASK)) {
+		request->sclp_req.status = SCLP_REQ_FAILED;
+		return -EIO;
+	}
+	request->sclp_req.command = SCLP_CMDW_WRITE_EVENT_DATA;
+	request->sclp_req.status = SCLP_REQ_FILLED;
+	request->sclp_req.callback = sclp_vt220_callback;
+	request->sclp_req.callback_data = (void *) request;
+
+	return sclp_add_request(&request->sclp_req);
+}
+
+/*
+ * Queue and emit current request.
+ */
+static void
+sclp_vt220_emit_current(void)
+{
+	unsigned long flags;
+	struct sclp_vt220_request *request;
+	struct sclp_vt220_sccb *sccb;
+
+	spin_lock_irqsave(&sclp_vt220_lock, flags);
+	if (sclp_vt220_current_request) {
+		sccb = (struct sclp_vt220_sccb *) 
+				sclp_vt220_current_request->sclp_req.sccb;
+		/* Only emit buffers with content */
+		if (sccb->header.length != sizeof(struct sclp_vt220_sccb)) {
+			list_add_tail(&sclp_vt220_current_request->list,
+				      &sclp_vt220_outqueue);
+			sclp_vt220_current_request = NULL;
+			if (timer_pending(&sclp_vt220_timer))
+				del_timer(&sclp_vt220_timer);
+		}
+		sclp_vt220_flush_later = 0;
+	}
+	if (sclp_vt220_queue_running || sclp_vt220_suspended)
+		goto out_unlock;
+	if (list_empty(&sclp_vt220_outqueue))
+		goto out_unlock;
+	request = list_first_entry(&sclp_vt220_outqueue,
+				   struct sclp_vt220_request, list);
+	sclp_vt220_queue_running = 1;
+	spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+
+	if (__sclp_vt220_emit(request))
+		sclp_vt220_process_queue(request);
+	return;
+out_unlock:
+	spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+}
+
+#define SCLP_NORMAL_WRITE	0x00
+
+/*
+ * Helper function to initialize a page with the sclp request structure.
+ */
+static struct sclp_vt220_request *
+sclp_vt220_initialize_page(void *page)
+{
+	struct sclp_vt220_request *request;
+	struct sclp_vt220_sccb *sccb;
+
+	/* Place request structure at end of page */
+	request = ((struct sclp_vt220_request *)
+			((addr_t) page + PAGE_SIZE)) - 1;
+	request->retry_count = 0;
+	request->sclp_req.sccb = page;
+	/* SCCB goes at start of page */
+	sccb = (struct sclp_vt220_sccb *) page;
+	memset((void *) sccb, 0, sizeof(struct sclp_vt220_sccb));
+	sccb->header.length = sizeof(struct sclp_vt220_sccb);
+	sccb->header.function_code = SCLP_NORMAL_WRITE;
+	sccb->header.response_code = 0x0000;
+	sccb->evbuf.type = EVTYP_VT220MSG;
+	sccb->evbuf.length = sizeof(struct evbuf_header);
+
+	return request;
+}
+
+static inline unsigned int
+sclp_vt220_space_left(struct sclp_vt220_request *request)
+{
+	struct sclp_vt220_sccb *sccb;
+	sccb = (struct sclp_vt220_sccb *) request->sclp_req.sccb;
+	return PAGE_SIZE - sizeof(struct sclp_vt220_request) -
+	       sccb->header.length;
+}
+
+static inline unsigned int
+sclp_vt220_chars_stored(struct sclp_vt220_request *request)
+{
+	struct sclp_vt220_sccb *sccb;
+	sccb = (struct sclp_vt220_sccb *) request->sclp_req.sccb;
+	return sccb->evbuf.length - sizeof(struct evbuf_header);
+}
+
+/*
+ * Add msg to buffer associated with request. Return the number of characters
+ * added.
+ */
+static int
+sclp_vt220_add_msg(struct sclp_vt220_request *request,
+		   const unsigned char *msg, int count, int convertlf)
+{
+	struct sclp_vt220_sccb *sccb;
+	void *buffer;
+	unsigned char c;
+	int from;
+	int to;
+
+	if (count > sclp_vt220_space_left(request))
+		count = sclp_vt220_space_left(request);
+	if (count <= 0)
+		return 0;
+
+	sccb = (struct sclp_vt220_sccb *) request->sclp_req.sccb;
+	buffer = (void *) ((addr_t) sccb + sccb->header.length);
+
+	if (convertlf) {
+		/* Perform Linefeed conversion (0x0a -> 0x0a 0x0d)*/
+		for (from=0, to=0;
+		     (from < count) && (to < sclp_vt220_space_left(request));
+		     from++) {
+			/* Retrieve character */
+			c = msg[from];
+			/* Perform conversion */
+			if (c == 0x0a) {
+				if (to + 1 < sclp_vt220_space_left(request)) {
+					((unsigned char *) buffer)[to++] = c;
+					((unsigned char *) buffer)[to++] = 0x0d;
+				} else
+					break;
+
+			} else
+				((unsigned char *) buffer)[to++] = c;
+		}
+		sccb->header.length += to;
+		sccb->evbuf.length += to;
+		return from;
+	} else {
+		memcpy(buffer, (const void *) msg, count);
+		sccb->header.length += count;
+		sccb->evbuf.length += count;
+		return count;
+	}
+}
+
+/*
+ * Emit buffer after having waited long enough for more data to arrive.
+ */
+static void
+sclp_vt220_timeout(unsigned long data)
+{
+	sclp_vt220_emit_current();
+}
+
+#define BUFFER_MAX_DELAY	HZ/20
+
+/* 
+ * Internal implementation of the write function. Write COUNT bytes of data
+ * from memory at BUF
+ * to the SCLP interface. In case that the data does not fit into the current
+ * write buffer, emit the current one and allocate a new one. If there are no
+ * more empty buffers available, wait until one gets emptied. If DO_SCHEDULE
+ * is non-zero, the buffer will be scheduled for emitting after a timeout -
+ * otherwise the user has to explicitly call the flush function.
+ * A non-zero CONVERTLF parameter indicates that 0x0a characters in the message
+ * buffer should be converted to 0x0a 0x0d. After completion, return the number
+ * of bytes written.
+ */
+static int
+__sclp_vt220_write(const unsigned char *buf, int count, int do_schedule,
+		   int convertlf, int may_fail)
+{
+	unsigned long flags;
+	void *page;
+	int written;
+	int overall_written;
+
+	if (count <= 0)
+		return 0;
+	overall_written = 0;
+	spin_lock_irqsave(&sclp_vt220_lock, flags);
+	do {
+		/* Create an sclp output buffer if none exists yet */
+		if (sclp_vt220_current_request == NULL) {
+			while (list_empty(&sclp_vt220_empty)) {
+				spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+				if (may_fail || sclp_vt220_suspended)
+					goto out;
+				else
+					sclp_sync_wait();
+				spin_lock_irqsave(&sclp_vt220_lock, flags);
+			}
+			page = (void *) sclp_vt220_empty.next;
+			list_del((struct list_head *) page);
+			sclp_vt220_current_request =
+				sclp_vt220_initialize_page(page);
+		}
+		/* Try to write the string to the current request buffer */
+		written = sclp_vt220_add_msg(sclp_vt220_current_request,
+					     buf, count, convertlf);
+		overall_written += written;
+		if (written == count)
+			break;
+		/*
+		 * Not all characters could be written to the current
+		 * output buffer. Emit the buffer, create a new buffer
+		 * and then output the rest of the string.
+		 */
+		spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+		sclp_vt220_emit_current();
+		spin_lock_irqsave(&sclp_vt220_lock, flags);
+		buf += written;
+		count -= written;
+	} while (count > 0);
+	/* Setup timer to output current console buffer after some time */
+	if (sclp_vt220_current_request != NULL &&
+	    !timer_pending(&sclp_vt220_timer) && do_schedule) {
+		sclp_vt220_timer.function = sclp_vt220_timeout;
+		sclp_vt220_timer.data = 0UL;
+		sclp_vt220_timer.expires = jiffies + BUFFER_MAX_DELAY;
+		add_timer(&sclp_vt220_timer);
+	}
+	spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+out:
+	return overall_written;
+}
+
+/*
+ * This routine is called by the kernel to write a series of
+ * characters to the tty device.  The characters may come from
+ * user space or kernel space.  This routine will return the
+ * number of characters actually accepted for writing.
+ */
+static int
+sclp_vt220_write(struct tty_struct *tty, const unsigned char *buf, int count)
+{
+	return __sclp_vt220_write(buf, count, 1, 0, 1);
+}
+
+#define SCLP_VT220_SESSION_ENDED	0x01
+#define	SCLP_VT220_SESSION_STARTED	0x80
+#define SCLP_VT220_SESSION_DATA		0x00
+
+/*
+ * Called by the SCLP to report incoming event buffers.
+ */
+static void
+sclp_vt220_receiver_fn(struct evbuf_header *evbuf)
+{
+	char *buffer;
+	unsigned int count;
+
+	/* Ignore input if device is not open */
+	if (sclp_vt220_tty == NULL)
+		return;
+
+	buffer = (char *) ((addr_t) evbuf + sizeof(struct evbuf_header));
+	count = evbuf->length - sizeof(struct evbuf_header);
+
+	switch (*buffer) {
+	case SCLP_VT220_SESSION_ENDED:
+	case SCLP_VT220_SESSION_STARTED:
+		break;
+	case SCLP_VT220_SESSION_DATA:
+		/* Send input to line discipline */
+		buffer++;
+		count--;
+		tty_insert_flip_string(sclp_vt220_tty, buffer, count);
+		tty_flip_buffer_push(sclp_vt220_tty);
+		break;
+	}
+}
+
+/*
+ * This routine is called when a particular tty device is opened.
+ */
+static int
+sclp_vt220_open(struct tty_struct *tty, struct file *filp)
+{
+	if (tty->count == 1) {
+		sclp_vt220_tty = tty;
+		tty->driver_data = kmalloc(SCLP_VT220_BUF_SIZE, GFP_KERNEL);
+		if (tty->driver_data == NULL)
+			return -ENOMEM;
+		tty->low_latency = 0;
+	}
+	return 0;
+}
+
+/*
+ * This routine is called when a particular tty device is closed.
+ */
+static void
+sclp_vt220_close(struct tty_struct *tty, struct file *filp)
+{
+	if (tty->count == 1) {
+		sclp_vt220_tty = NULL;
+		kfree(tty->driver_data);
+		tty->driver_data = NULL;
+	}
+}
+
+/*
+ * This routine is called by the kernel to write a single
+ * character to the tty device.  If the kernel uses this routine,
+ * it must call the flush_chars() routine (if defined) when it is
+ * done stuffing characters into the driver.
+ */
+static int
+sclp_vt220_put_char(struct tty_struct *tty, unsigned char ch)
+{
+	return __sclp_vt220_write(&ch, 1, 0, 0, 1);
+}
+
+/*
+ * This routine is called by the kernel after it has written a
+ * series of characters to the tty device using put_char().  
+ */
+static void
+sclp_vt220_flush_chars(struct tty_struct *tty)
+{
+	if (!sclp_vt220_queue_running)
+		sclp_vt220_emit_current();
+	else
+		sclp_vt220_flush_later = 1;
+}
+
+/*
+ * This routine returns the numbers of characters the tty driver
+ * will accept for queuing to be written.  This number is subject
+ * to change as output buffers get emptied, or if the output flow
+ * control is acted.
+ */
+static int
+sclp_vt220_write_room(struct tty_struct *tty)
+{
+	unsigned long flags;
+	struct list_head *l;
+	int count;
+
+	spin_lock_irqsave(&sclp_vt220_lock, flags);
+	count = 0;
+	if (sclp_vt220_current_request != NULL)
+		count = sclp_vt220_space_left(sclp_vt220_current_request);
+	list_for_each(l, &sclp_vt220_empty)
+		count += SCLP_VT220_MAX_CHARS_PER_BUFFER;
+	spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+	return count;
+}
+
+/*
+ * Return number of buffered chars.
+ */
+static int
+sclp_vt220_chars_in_buffer(struct tty_struct *tty)
+{
+	unsigned long flags;
+	struct list_head *l;
+	struct sclp_vt220_request *r;
+	int count;
+
+	spin_lock_irqsave(&sclp_vt220_lock, flags);
+	count = 0;
+	if (sclp_vt220_current_request != NULL)
+		count = sclp_vt220_chars_stored(sclp_vt220_current_request);
+	list_for_each(l, &sclp_vt220_outqueue) {
+		r = list_entry(l, struct sclp_vt220_request, list);
+		count += sclp_vt220_chars_stored(r);
+	}
+	spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+	return count;
+}
+
+/*
+ * Pass on all buffers to the hardware. Return only when there are no more
+ * buffers pending.
+ */
+static void
+sclp_vt220_flush_buffer(struct tty_struct *tty)
+{
+	sclp_vt220_emit_current();
+}
+
+/* Release allocated pages. */
+static void __init __sclp_vt220_free_pages(void)
+{
+	struct list_head *page, *p;
+
+	list_for_each_safe(page, p, &sclp_vt220_empty) {
+		list_del(page);
+		free_page((unsigned long) page);
+	}
+}
+
+/* Release memory and unregister from sclp core. Controlled by init counting -
+ * only the last invoker will actually perform these actions. */
+static void __init __sclp_vt220_cleanup(void)
+{
+	sclp_vt220_init_count--;
+	if (sclp_vt220_init_count != 0)
+		return;
+	sclp_unregister(&sclp_vt220_register);
+	__sclp_vt220_free_pages();
+}
+
+/* Allocate buffer pages and register with sclp core. Controlled by init
+ * counting - only the first invoker will actually perform these actions. */
+static int __init __sclp_vt220_init(int num_pages)
+{
+	void *page;
+	int i;
+	int rc;
+
+	sclp_vt220_init_count++;
+	if (sclp_vt220_init_count != 1)
+		return 0;
+	spin_lock_init(&sclp_vt220_lock);
+	INIT_LIST_HEAD(&sclp_vt220_empty);
+	INIT_LIST_HEAD(&sclp_vt220_outqueue);
+	init_timer(&sclp_vt220_timer);
+	sclp_vt220_current_request = NULL;
+	sclp_vt220_buffered_chars = 0;
+	sclp_vt220_tty = NULL;
+	sclp_vt220_flush_later = 0;
+
+	/* Allocate pages for output buffering */
+	rc = -ENOMEM;
+	for (i = 0; i < num_pages; i++) {
+		page = (void *) get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		if (!page)
+			goto out;
+		list_add_tail(page, &sclp_vt220_empty);
+	}
+	rc = sclp_register(&sclp_vt220_register);
+out:
+	if (rc) {
+		__sclp_vt220_free_pages();
+		sclp_vt220_init_count--;
+	}
+	return rc;
+}
+
+static const struct tty_operations sclp_vt220_ops = {
+	.open = sclp_vt220_open,
+	.close = sclp_vt220_close,
+	.write = sclp_vt220_write,
+	.put_char = sclp_vt220_put_char,
+	.flush_chars = sclp_vt220_flush_chars,
+	.write_room = sclp_vt220_write_room,
+	.chars_in_buffer = sclp_vt220_chars_in_buffer,
+	.flush_buffer = sclp_vt220_flush_buffer,
+};
+
+/*
+ * Register driver with SCLP and Linux and initialize internal tty structures.
+ */
+static int __init sclp_vt220_tty_init(void)
+{
+	struct tty_driver *driver;
+	int rc;
+
+	/* Note: we're not testing for CONSOLE_IS_SCLP here to preserve
+	 * symmetry between VM and LPAR systems regarding ttyS1. */
+	driver = alloc_tty_driver(1);
+	if (!driver)
+		return -ENOMEM;
+	rc = __sclp_vt220_init(MAX_KMEM_PAGES);
+	if (rc)
+		goto out_driver;
+
+	driver->owner = THIS_MODULE;
+	driver->driver_name = SCLP_VT220_DRIVER_NAME;
+	driver->name = SCLP_VT220_DEVICE_NAME;
+	driver->major = SCLP_VT220_MAJOR;
+	driver->minor_start = SCLP_VT220_MINOR;
+	driver->type = TTY_DRIVER_TYPE_SYSTEM;
+	driver->subtype = SYSTEM_TYPE_TTY;
+	driver->init_termios = tty_std_termios;
+	driver->flags = TTY_DRIVER_REAL_RAW;
+	tty_set_operations(driver, &sclp_vt220_ops);
+
+	rc = tty_register_driver(driver);
+	if (rc)
+		goto out_init;
+	sclp_vt220_driver = driver;
+	return 0;
+
+out_init:
+	__sclp_vt220_cleanup();
+out_driver:
+	put_tty_driver(driver);
+	return rc;
+}
+__initcall(sclp_vt220_tty_init);
+
+static void __sclp_vt220_flush_buffer(void)
+{
+	unsigned long flags;
+
+	sclp_vt220_emit_current();
+	spin_lock_irqsave(&sclp_vt220_lock, flags);
+	if (timer_pending(&sclp_vt220_timer))
+		del_timer(&sclp_vt220_timer);
+	while (sclp_vt220_queue_running) {
+		spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+		sclp_sync_wait();
+		spin_lock_irqsave(&sclp_vt220_lock, flags);
+	}
+	spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+}
+
+/*
+ * Resume console: If there are cached messages, emit them.
+ */
+static void sclp_vt220_resume(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sclp_vt220_lock, flags);
+	sclp_vt220_suspended = 0;
+	spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+	sclp_vt220_emit_current();
+}
+
+/*
+ * Suspend console: Set suspend flag and flush console
+ */
+static void sclp_vt220_suspend(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sclp_vt220_lock, flags);
+	sclp_vt220_suspended = 1;
+	spin_unlock_irqrestore(&sclp_vt220_lock, flags);
+	__sclp_vt220_flush_buffer();
+}
+
+static void sclp_vt220_pm_event_fn(struct sclp_register *reg,
+				   enum sclp_pm_event sclp_pm_event)
+{
+	switch (sclp_pm_event) {
+	case SCLP_PM_EVENT_FREEZE:
+		sclp_vt220_suspend();
+		break;
+	case SCLP_PM_EVENT_RESTORE:
+	case SCLP_PM_EVENT_THAW:
+		sclp_vt220_resume();
+		break;
+	}
+}
+
+#ifdef CONFIG_SCLP_VT220_CONSOLE
+
+static void
+sclp_vt220_con_write(struct console *con, const char *buf, unsigned int count)
+{
+	__sclp_vt220_write((const unsigned char *) buf, count, 1, 1, 0);
+}
+
+static struct tty_driver *
+sclp_vt220_con_device(struct console *c, int *index)
+{
+	*index = 0;
+	return sclp_vt220_driver;
+}
+
+static int
+sclp_vt220_notify(struct notifier_block *self,
+			  unsigned long event, void *data)
+{
+	__sclp_vt220_flush_buffer();
+	return NOTIFY_OK;
+}
+
+static struct notifier_block on_panic_nb = {
+	.notifier_call = sclp_vt220_notify,
+	.priority = 1,
+};
+
+static struct notifier_block on_reboot_nb = {
+	.notifier_call = sclp_vt220_notify,
+	.priority = 1,
+};
+
+/* Structure needed to register with printk */
+static struct console sclp_vt220_console =
+{
+	.name = SCLP_VT220_CONSOLE_NAME,
+	.write = sclp_vt220_con_write,
+	.device = sclp_vt220_con_device,
+	.flags = CON_PRINTBUFFER,
+	.index = SCLP_VT220_CONSOLE_INDEX
+};
+
+static int __init
+sclp_vt220_con_init(void)
+{
+	int rc;
+
+	if (!CONSOLE_IS_SCLP)
+		return 0;
+	rc = __sclp_vt220_init(MAX_CONSOLE_PAGES);
+	if (rc)
+		return rc;
+	/* Attach linux console */
+	atomic_notifier_chain_register(&panic_notifier_list, &on_panic_nb);
+	register_reboot_notifier(&on_reboot_nb);
+	register_console(&sclp_vt220_console);
+	return 0;
+}
+
+console_initcall(sclp_vt220_con_init);
+#endif /* CONFIG_SCLP_VT220_CONSOLE */
+

kernel/drivers/s390/char/tape.h

@@ -0,0 +1,403 @@
+/*
+ *  drivers/s390/char/tape.h
+ *    tape device driver for 3480/3490E/3590 tapes.
+ *
+ *  S390 and zSeries version
+ *    Copyright IBM Corp. 2001, 2009
+ *    Author(s): Carsten Otte <cotte@de.ibm.com>
+ *		 Tuan Ngo-Anh <ngoanh@de.ibm.com>
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *		 Stefan Bader <shbader@de.ibm.com>
+ */
+
+#ifndef _TAPE_H
+#define _TAPE_H
+
+#include <asm/ccwdev.h>
+#include <asm/debug.h>
+#include <asm/idals.h>
+#include <linux/blkdev.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mtio.h>
+#include <linux/interrupt.h>
+#include <linux/workqueue.h>
+
+struct gendisk;
+
+/*
+ * Define DBF_LIKE_HELL for lots of messages in the debug feature.
+ */
+#define DBF_LIKE_HELL
+#ifdef  DBF_LIKE_HELL
+#define DBF_LH(level, str, ...) \
+do { \
+	debug_sprintf_event(TAPE_DBF_AREA, level, str, ## __VA_ARGS__); \
+} while (0)
+#else
+#define DBF_LH(level, str, ...) do {} while(0)
+#endif
+
+/*
+ * macros s390 debug feature (dbf)
+ */
+#define DBF_EVENT(d_level, d_str...) \
+do { \
+	debug_sprintf_event(TAPE_DBF_AREA, d_level, d_str); \
+} while (0)
+
+#define DBF_EXCEPTION(d_level, d_str...) \
+do { \
+	debug_sprintf_exception(TAPE_DBF_AREA, d_level, d_str); \
+} while (0)
+
+#define TAPE_VERSION_MAJOR 2
+#define TAPE_VERSION_MINOR 0
+#define TAPE_MAGIC "tape"
+
+#define TAPE_MINORS_PER_DEV 2	    /* two minors per device */
+#define TAPEBLOCK_HSEC_SIZE	2048
+#define TAPEBLOCK_HSEC_S2B	2
+#define TAPEBLOCK_RETRIES	5
+
+enum tape_medium_state {
+	MS_UNKNOWN,
+	MS_LOADED,
+	MS_UNLOADED,
+	MS_SIZE
+};
+
+enum tape_state {
+	TS_UNUSED=0,
+	TS_IN_USE,
+	TS_BLKUSE,
+	TS_INIT,
+	TS_NOT_OPER,
+	TS_SIZE
+};
+
+enum tape_op {
+	TO_BLOCK,	/* Block read */
+	TO_BSB,		/* Backward space block */
+	TO_BSF,		/* Backward space filemark */
+	TO_DSE,		/* Data security erase */
+	TO_FSB,		/* Forward space block */
+	TO_FSF,		/* Forward space filemark */
+	TO_LBL,		/* Locate block label */
+	TO_NOP,		/* No operation */
+	TO_RBA,		/* Read backward */
+	TO_RBI,		/* Read block information */
+	TO_RFO,		/* Read forward */
+	TO_REW,		/* Rewind tape */
+	TO_RUN,		/* Rewind and unload tape */
+	TO_WRI,		/* Write block */
+	TO_WTM,		/* Write tape mark */
+	TO_MSEN,	/* Medium sense */
+	TO_LOAD,	/* Load tape */
+	TO_READ_CONFIG, /* Read configuration data */
+	TO_READ_ATTMSG, /* Read attention message */
+	TO_DIS,		/* Tape display */
+	TO_ASSIGN,	/* Assign tape to channel path */
+	TO_UNASSIGN,	/* Unassign tape from channel path */
+	TO_CRYPT_ON,	/* Enable encrpytion */
+	TO_CRYPT_OFF,	/* Disable encrpytion */
+	TO_KEKL_SET,	/* Set KEK label */
+	TO_KEKL_QUERY,	/* Query KEK label */
+	TO_RDC,		/* Read device characteristics */
+	TO_SIZE,	/* #entries in tape_op_t */
+};
+
+/* Forward declaration */
+struct tape_device;
+
+/* tape_request->status can be: */
+enum tape_request_status {
+	TAPE_REQUEST_INIT,	/* request is ready to be processed */
+	TAPE_REQUEST_QUEUED,	/* request is queued to be processed */
+	TAPE_REQUEST_IN_IO,	/* request is currently in IO */
+	TAPE_REQUEST_DONE,	/* request is completed. */
+	TAPE_REQUEST_CANCEL,	/* request should be canceled. */
+	TAPE_REQUEST_LONG_BUSY, /* request has to be restarted after long busy */
+};
+
+/* Tape CCW request */
+struct tape_request {
+	struct list_head list;		/* list head for request queueing. */
+	struct tape_device *device;	/* tape device of this request */
+	struct ccw1 *cpaddr;		/* address of the channel program. */
+	void *cpdata;			/* pointer to ccw data. */
+	enum tape_request_status status;/* status of this request */
+	int options;			/* options for execution. */
+	int retries;			/* retry counter for error recovery. */
+	int rescnt;			/* residual count from devstat. */
+
+	/* Callback for delivering final status. */
+	void (*callback)(struct tape_request *, void *);
+	void *callback_data;
+
+	enum tape_op op;
+	int rc;
+};
+
+/* Function type for magnetic tape commands */
+typedef int (*tape_mtop_fn)(struct tape_device *, int);
+
+/* Size of the array containing the mtops for a discipline */
+#define TAPE_NR_MTOPS (MTMKPART+1)
+
+/* Tape Discipline */
+struct tape_discipline {
+	struct module *owner;
+	int  (*setup_device)(struct tape_device *);
+	void (*cleanup_device)(struct tape_device *);
+	int (*irq)(struct tape_device *, struct tape_request *, struct irb *);
+	struct tape_request *(*read_block)(struct tape_device *, size_t);
+	struct tape_request *(*write_block)(struct tape_device *, size_t);
+	void (*process_eov)(struct tape_device*);
+#ifdef CONFIG_S390_TAPE_BLOCK
+	/* Block device stuff. */
+	struct tape_request *(*bread)(struct tape_device *, struct request *);
+	void (*check_locate)(struct tape_device *, struct tape_request *);
+	void (*free_bread)(struct tape_request *);
+#endif
+	/* ioctl function for additional ioctls. */
+	int (*ioctl_fn)(struct tape_device *, unsigned int, unsigned long);
+	/* Array of tape commands with TAPE_NR_MTOPS entries */
+	tape_mtop_fn *mtop_array;
+};
+
+/*
+ * The discipline irq function either returns an error code (<0) which
+ * means that the request has failed with an error or one of the following:
+ */
+#define TAPE_IO_SUCCESS		0	/* request successful */
+#define TAPE_IO_PENDING		1	/* request still running */
+#define TAPE_IO_RETRY		2	/* retry to current request */
+#define TAPE_IO_STOP		3	/* stop the running request */
+#define TAPE_IO_LONG_BUSY	4	/* delay the running request */
+
+/* Char Frontend Data */
+struct tape_char_data {
+	struct idal_buffer *idal_buf;	/* idal buffer for user char data */
+	int block_size;			/*   of size block_size. */
+};
+
+#ifdef CONFIG_S390_TAPE_BLOCK
+/* Block Frontend Data */
+struct tape_blk_data
+{
+	struct tape_device *	device;
+	/* Block device request queue. */
+	struct request_queue *	request_queue;
+	spinlock_t		request_queue_lock;
+
+	/* Task to move entries from block request to CCS request queue. */
+	struct work_struct	requeue_task;
+	atomic_t		requeue_scheduled;
+
+	/* Current position on the tape. */
+	long			block_position;
+	int			medium_changed;
+	struct gendisk *	disk;
+};
+#endif
+
+/* Tape Info */
+struct tape_device {
+	/* entry in tape_device_list */
+	struct list_head		node;
+
+	int				cdev_id;
+	struct ccw_device *		cdev;
+	struct tape_class_device *	nt;
+	struct tape_class_device *	rt;
+
+	/* Device discipline information. */
+	struct tape_discipline *	discipline;
+	void *				discdata;
+
+	/* Generic status flags */
+	long				tape_generic_status;
+
+	/* Device state information. */
+	wait_queue_head_t		state_change_wq;
+	enum tape_state			tape_state;
+	enum tape_medium_state		medium_state;
+	unsigned char *			modeset_byte;
+
+	/* Reference count. */
+	atomic_t			ref_count;
+
+	/* Request queue. */
+	struct list_head		req_queue;
+
+	/* Request wait queue. */
+	wait_queue_head_t		wait_queue;
+
+	/* Each tape device has (currently) two minor numbers. */
+	int				first_minor;
+
+	/* Number of tapemarks required for correct termination. */
+	int				required_tapemarks;
+
+	/* Block ID of the BOF */
+	unsigned int			bof;
+
+	/* Character device frontend data */
+	struct tape_char_data		char_data;
+#ifdef CONFIG_S390_TAPE_BLOCK
+	/* Block dev frontend data */
+	struct tape_blk_data		blk_data;
+#endif
+
+	/* Function to start or stop the next request later. */
+	struct delayed_work		tape_dnr;
+
+	/* Timer for long busy */
+	struct timer_list		lb_timeout;
+
+};
+
+/* Externals from tape_core.c */
+extern struct tape_request *tape_alloc_request(int cplength, int datasize);
+extern void tape_free_request(struct tape_request *);
+extern int tape_do_io(struct tape_device *, struct tape_request *);
+extern int tape_do_io_async(struct tape_device *, struct tape_request *);
+extern int tape_do_io_interruptible(struct tape_device *, struct tape_request *);
+extern int tape_cancel_io(struct tape_device *, struct tape_request *);
+void tape_hotplug_event(struct tape_device *, int major, int action);
+
+static inline int
+tape_do_io_free(struct tape_device *device, struct tape_request *request)
+{
+	int rc;
+
+	rc = tape_do_io(device, request);
+	tape_free_request(request);
+	return rc;
+}
+
+extern int tape_oper_handler(int irq, int status);
+extern void tape_noper_handler(int irq, int status);
+extern int tape_open(struct tape_device *);
+extern int tape_release(struct tape_device *);
+extern int tape_mtop(struct tape_device *, int, int);
+extern void tape_state_set(struct tape_device *, enum tape_state);
+
+extern int tape_generic_online(struct tape_device *, struct tape_discipline *);
+extern int tape_generic_offline(struct ccw_device *);
+extern int tape_generic_pm_suspend(struct ccw_device *);
+
+/* Externals from tape_devmap.c */
+extern int tape_generic_probe(struct ccw_device *);
+extern void tape_generic_remove(struct ccw_device *);
+
+extern struct tape_device *tape_get_device(int devindex);
+extern struct tape_device *tape_get_device_reference(struct tape_device *);
+extern struct tape_device *tape_put_device(struct tape_device *);
+
+/* Externals from tape_char.c */
+extern int tapechar_init(void);
+extern void tapechar_exit(void);
+extern int  tapechar_setup_device(struct tape_device *);
+extern void tapechar_cleanup_device(struct tape_device *);
+
+/* Externals from tape_block.c */
+#ifdef CONFIG_S390_TAPE_BLOCK
+extern int tapeblock_init (void);
+extern void tapeblock_exit(void);
+extern int tapeblock_setup_device(struct tape_device *);
+extern void tapeblock_cleanup_device(struct tape_device *);
+#else
+static inline int tapeblock_init (void) {return 0;}
+static inline void tapeblock_exit (void) {;}
+static inline int tapeblock_setup_device(struct tape_device *t) {return 0;}
+static inline void tapeblock_cleanup_device (struct tape_device *t) {;}
+#endif
+
+/* tape initialisation functions */
+#ifdef CONFIG_PROC_FS
+extern void tape_proc_init (void);
+extern void tape_proc_cleanup (void);
+#else
+static inline void tape_proc_init (void) {;}
+static inline void tape_proc_cleanup (void) {;}
+#endif
+
+/* a function for dumping device sense info */
+extern void tape_dump_sense_dbf(struct tape_device *, struct tape_request *,
+				struct irb *);
+
+/* functions for handling the status of a device */
+extern void tape_med_state_set(struct tape_device *, enum tape_medium_state);
+
+/* The debug area */
+extern debug_info_t *TAPE_DBF_AREA;
+
+/* functions for building ccws */
+static inline struct ccw1 *
+tape_ccw_cc(struct ccw1 *ccw, __u8 cmd_code, __u16 memsize, void *cda)
+{
+	ccw->cmd_code = cmd_code;
+	ccw->flags = CCW_FLAG_CC;
+	ccw->count = memsize;
+	ccw->cda = (__u32)(addr_t) cda;
+	return ccw + 1;
+}
+
+static inline struct ccw1 *
+tape_ccw_end(struct ccw1 *ccw, __u8 cmd_code, __u16 memsize, void *cda)
+{
+	ccw->cmd_code = cmd_code;
+	ccw->flags = 0;
+	ccw->count = memsize;
+	ccw->cda = (__u32)(addr_t) cda;
+	return ccw + 1;
+}
+
+static inline struct ccw1 *
+tape_ccw_cmd(struct ccw1 *ccw, __u8 cmd_code)
+{
+	ccw->cmd_code = cmd_code;
+	ccw->flags = 0;
+	ccw->count = 0;
+	ccw->cda = (__u32)(addr_t) &ccw->cmd_code;
+	return ccw + 1;
+}
+
+static inline struct ccw1 *
+tape_ccw_repeat(struct ccw1 *ccw, __u8 cmd_code, int count)
+{
+	while (count-- > 0) {
+		ccw->cmd_code = cmd_code;
+		ccw->flags = CCW_FLAG_CC;
+		ccw->count = 0;
+		ccw->cda = (__u32)(addr_t) &ccw->cmd_code;
+		ccw++;
+	}
+	return ccw;
+}
+
+static inline struct ccw1 *
+tape_ccw_cc_idal(struct ccw1 *ccw, __u8 cmd_code, struct idal_buffer *idal)
+{
+	ccw->cmd_code = cmd_code;
+	ccw->flags    = CCW_FLAG_CC;
+	idal_buffer_set_cda(idal, ccw);
+	return ccw++;
+}
+
+static inline struct ccw1 *
+tape_ccw_end_idal(struct ccw1 *ccw, __u8 cmd_code, struct idal_buffer *idal)
+{
+	ccw->cmd_code = cmd_code;
+	ccw->flags    = 0;
+	idal_buffer_set_cda(idal, ccw);
+	return ccw++;
+}
+
+/* Global vars */
+extern const char *tape_state_verbose[];
+extern const char *tape_op_verbose[];
+
+#endif /* for ifdef tape.h */

kernel/drivers/s390/char/tape_3590.h

@@ -0,0 +1,175 @@
+/*
+ *  drivers/s390/char/tape_3590.h
+ *    tape device discipline for 3590 tapes.
+ *
+ *    Copyright IBM Corp. 2001,2006
+ *    Author(s): Stefan Bader <shbader@de.ibm.com>
+ *		 Michael Holzheu <holzheu@de.ibm.com>
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#ifndef _TAPE_3590_H
+#define _TAPE_3590_H
+
+#define MEDIUM_SENSE	0xc2
+#define READ_PREVIOUS	0x0a
+#define MODE_SENSE	0xcf
+#define PERFORM_SS_FUNC 0x77
+#define READ_SS_DATA	0x3e
+
+#define PREP_RD_SS_DATA 0x18
+#define RD_ATTMSG	0x3
+
+#define SENSE_BRA_PER  0
+#define SENSE_BRA_CONT 1
+#define SENSE_BRA_RE   2
+#define SENSE_BRA_DRE  3
+
+#define SENSE_FMT_LIBRARY	0x23
+#define SENSE_FMT_UNSOLICITED	0x40
+#define SENSE_FMT_COMMAND_REJ	0x41
+#define SENSE_FMT_COMMAND_EXEC0 0x50
+#define SENSE_FMT_COMMAND_EXEC1 0x51
+#define SENSE_FMT_EVENT0	0x60
+#define SENSE_FMT_EVENT1	0x61
+#define SENSE_FMT_MIM		0x70
+#define SENSE_FMT_SIM		0x71
+
+#define MSENSE_UNASSOCIATED	 0x00
+#define MSENSE_ASSOCIATED_MOUNT	 0x01
+#define MSENSE_ASSOCIATED_UMOUNT 0x02
+#define MSENSE_CRYPT_MASK	 0x00000010
+
+#define TAPE_3590_MAX_MSG	 0xb0
+
+/* Datatypes */
+
+struct tape_3590_disc_data {
+	struct tape390_crypt_info crypt_info;
+	int read_back_op;
+};
+
+#define TAPE_3590_CRYPT_INFO(device) \
+	((struct tape_3590_disc_data*)(device->discdata))->crypt_info
+#define TAPE_3590_READ_BACK_OP(device) \
+	((struct tape_3590_disc_data*)(device->discdata))->read_back_op
+
+struct tape_3590_sense {
+
+	unsigned int command_rej:1;
+	unsigned int interv_req:1;
+	unsigned int bus_out_check:1;
+	unsigned int eq_check:1;
+	unsigned int data_check:1;
+	unsigned int overrun:1;
+	unsigned int def_unit_check:1;
+	unsigned int assgnd_elsew:1;
+
+	unsigned int locate_fail:1;
+	unsigned int inst_online:1;
+	unsigned int reserved:1;
+	unsigned int blk_seq_err:1;
+	unsigned int begin_part:1;
+	unsigned int wr_mode:1;
+	unsigned int wr_prot:1;
+	unsigned int not_cap:1;
+
+	unsigned int bra:2;
+	unsigned int lc:3;
+	unsigned int vlf_active:1;
+	unsigned int stm:1;
+	unsigned int med_pos:1;
+
+	unsigned int rac:8;
+
+	unsigned int rc_rqc:16;
+
+	unsigned int mc:8;
+
+	unsigned int sense_fmt:8;
+
+	union {
+		struct {
+			unsigned int emc:4;
+			unsigned int smc:4;
+			unsigned int sev:2;
+			unsigned int reserved:6;
+			unsigned int md:8;
+			unsigned int refcode:8;
+			unsigned int mid:16;
+			unsigned int mp:16;
+			unsigned char volid[6];
+			unsigned int fid:8;
+		} f70;
+		struct {
+			unsigned int emc:4;
+			unsigned int smc:4;
+			unsigned int sev:2;
+			unsigned int reserved1:5;
+			unsigned int mdf:1;
+			unsigned char md[3];
+			unsigned int simid:8;
+			unsigned int uid:16;
+			unsigned int refcode1:16;
+			unsigned int refcode2:16;
+			unsigned int refcode3:16;
+			unsigned int reserved2:8;
+		} f71;
+		unsigned char data[14];
+	} fmt;
+	unsigned char pad[10];
+
+} __attribute__ ((packed));
+
+struct tape_3590_med_sense {
+	unsigned int macst:4;
+	unsigned int masst:4;
+	char pad1[7];
+	unsigned int flags;
+	char pad2[116];
+} __attribute__ ((packed));
+
+struct tape_3590_rdc_data {
+	char data[64];
+} __attribute__ ((packed));
+
+/* Datastructures for 3592 encryption support */
+
+struct tape3592_kekl {
+	__u8 flags;
+	char label[64];
+} __attribute__ ((packed));
+
+struct tape3592_kekl_pair {
+	__u8 count;
+	struct tape3592_kekl kekl[2];
+} __attribute__ ((packed));
+
+struct tape3592_kekl_query_data {
+	__u16 len;
+	__u8  fmt;
+	__u8  mc;
+	__u32 id;
+	__u8  flags;
+	struct tape3592_kekl_pair kekls;
+	char reserved[116];
+} __attribute__ ((packed));
+
+struct tape3592_kekl_query_order {
+	__u8 code;
+	__u8 flags;
+	char reserved1[2];
+	__u8 max_count;
+	char reserved2[35];
+} __attribute__ ((packed));
+
+struct tape3592_kekl_set_order {
+	__u8 code;
+	__u8 flags;
+	char reserved1[2];
+	__u8 op;
+	struct tape3592_kekl_pair kekls;
+	char reserved2[120];
+} __attribute__ ((packed));
+
+#endif /* _TAPE_3590_H */

kernel/drivers/s390/char/tape_block.c

@@ -0,0 +1,477 @@
+/*
+ *  drivers/s390/char/tape_block.c
+ *    block device frontend for tape device driver
+ *
+ *  S390 and zSeries version
+ *    Copyright (C) 2001,2003 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Author(s): Carsten Otte <cotte@de.ibm.com>
+ *		 Tuan Ngo-Anh <ngoanh@de.ibm.com>
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *		 Stefan Bader <shbader@de.ibm.com>
+ */
+
+#define KMSG_COMPONENT "tape"
+
+#include <linux/fs.h>
+#include <linux/module.h>
+#include <linux/blkdev.h>
+#include <linux/interrupt.h>
+#include <linux/buffer_head.h>
+#include <linux/kernel.h>
+
+#include <asm/debug.h>
+
+#define TAPE_DBF_AREA	tape_core_dbf
+
+#include "tape.h"
+
+#define TAPEBLOCK_MAX_SEC	100
+#define TAPEBLOCK_MIN_REQUEUE	3
+
+/*
+ * 2003/11/25  Stefan Bader <shbader@de.ibm.com>
+ *
+ * In 2.5/2.6 the block device request function is very likely to be called
+ * with disabled interrupts (e.g. generic_unplug_device). So the driver can't
+ * just call any function that tries to allocate CCW requests from that con-
+ * text since it might sleep. There are two choices to work around this:
+ *	a) do not allocate with kmalloc but use its own memory pool
+ *      b) take requests from the queue outside that context, knowing that
+ *         allocation might sleep
+ */
+
+/*
+ * file operation structure for tape block frontend
+ */
+static int tapeblock_open(struct block_device *, fmode_t);
+static int tapeblock_release(struct gendisk *, fmode_t);
+static int tapeblock_ioctl(struct block_device *, fmode_t, unsigned int,
+				unsigned long);
+static int tapeblock_medium_changed(struct gendisk *);
+static int tapeblock_revalidate_disk(struct gendisk *);
+
+static const struct block_device_operations tapeblock_fops = {
+	.owner		 = THIS_MODULE,
+	.open		 = tapeblock_open,
+	.release	 = tapeblock_release,
+	.locked_ioctl           = tapeblock_ioctl,
+	.media_changed   = tapeblock_medium_changed,
+	.revalidate_disk = tapeblock_revalidate_disk,
+};
+
+static int tapeblock_major = 0;
+
+static void
+tapeblock_trigger_requeue(struct tape_device *device)
+{
+	/* Protect against rescheduling. */
+	if (atomic_cmpxchg(&device->blk_data.requeue_scheduled, 0, 1) != 0)
+		return;
+	schedule_work(&device->blk_data.requeue_task);
+}
+
+/*
+ * Post finished request.
+ */
+static void
+__tapeblock_end_request(struct tape_request *ccw_req, void *data)
+{
+	struct tape_device *device;
+	struct request *req;
+
+	DBF_LH(6, "__tapeblock_end_request()\n");
+
+	device = ccw_req->device;
+	req = (struct request *) data;
+	blk_end_request_all(req, (ccw_req->rc == 0) ? 0 : -EIO);
+	if (ccw_req->rc == 0)
+		/* Update position. */
+		device->blk_data.block_position =
+		  (blk_rq_pos(req) + blk_rq_sectors(req)) >> TAPEBLOCK_HSEC_S2B;
+	else
+		/* We lost the position information due to an error. */
+		device->blk_data.block_position = -1;
+	device->discipline->free_bread(ccw_req);
+	if (!list_empty(&device->req_queue) ||
+	    blk_peek_request(device->blk_data.request_queue))
+		tapeblock_trigger_requeue(device);
+}
+
+/*
+ * Feed the tape device CCW queue with requests supplied in a list.
+ */
+static int
+tapeblock_start_request(struct tape_device *device, struct request *req)
+{
+	struct tape_request *	ccw_req;
+	int			rc;
+
+	DBF_LH(6, "tapeblock_start_request(%p, %p)\n", device, req);
+
+	ccw_req = device->discipline->bread(device, req);
+	if (IS_ERR(ccw_req)) {
+		DBF_EVENT(1, "TBLOCK: bread failed\n");
+		blk_end_request_all(req, -EIO);
+		return PTR_ERR(ccw_req);
+	}
+	ccw_req->callback = __tapeblock_end_request;
+	ccw_req->callback_data = (void *) req;
+	ccw_req->retries = TAPEBLOCK_RETRIES;
+
+	rc = tape_do_io_async(device, ccw_req);
+	if (rc) {
+		/*
+		 * Start/enqueueing failed. No retries in
+		 * this case.
+		 */
+		blk_end_request_all(req, -EIO);
+		device->discipline->free_bread(ccw_req);
+	}
+
+	return rc;
+}
+
+/*
+ * Move requests from the block device request queue to the tape device ccw
+ * queue.
+ */
+static void
+tapeblock_requeue(struct work_struct *work) {
+	struct tape_blk_data *	blkdat;
+	struct tape_device *	device;
+	struct request_queue *	queue;
+	int			nr_queued;
+	struct request *	req;
+	struct list_head *	l;
+	int			rc;
+
+	blkdat = container_of(work, struct tape_blk_data, requeue_task);
+	device = blkdat->device;
+	if (!device)
+		return;
+
+	spin_lock_irq(get_ccwdev_lock(device->cdev));
+	queue  = device->blk_data.request_queue;
+
+	/* Count number of requests on ccw queue. */
+	nr_queued = 0;
+	list_for_each(l, &device->req_queue)
+		nr_queued++;
+	spin_unlock(get_ccwdev_lock(device->cdev));
+
+	spin_lock_irq(&device->blk_data.request_queue_lock);
+	while (
+		!blk_queue_plugged(queue) &&
+		blk_peek_request(queue) &&
+		nr_queued < TAPEBLOCK_MIN_REQUEUE
+	) {
+		req = blk_fetch_request(queue);
+		if (rq_data_dir(req) == WRITE) {
+			DBF_EVENT(1, "TBLOCK: Rejecting write request\n");
+			spin_unlock_irq(&device->blk_data.request_queue_lock);
+			blk_end_request_all(req, -EIO);
+			spin_lock_irq(&device->blk_data.request_queue_lock);
+			continue;
+		}
+		nr_queued++;
+		spin_unlock_irq(&device->blk_data.request_queue_lock);
+		rc = tapeblock_start_request(device, req);
+		spin_lock_irq(&device->blk_data.request_queue_lock);
+	}
+	spin_unlock_irq(&device->blk_data.request_queue_lock);
+	atomic_set(&device->blk_data.requeue_scheduled, 0);
+}
+
+/*
+ * Tape request queue function. Called from ll_rw_blk.c
+ */
+static void
+tapeblock_request_fn(struct request_queue *queue)
+{
+	struct tape_device *device;
+
+	device = (struct tape_device *) queue->queuedata;
+	DBF_LH(6, "tapeblock_request_fn(device=%p)\n", device);
+	BUG_ON(device == NULL);
+	tapeblock_trigger_requeue(device);
+}
+
+/*
+ * This function is called for every new tapedevice
+ */
+int
+tapeblock_setup_device(struct tape_device * device)
+{
+	struct tape_blk_data *	blkdat;
+	struct gendisk *	disk;
+	int			rc;
+
+	blkdat = &device->blk_data;
+	blkdat->device = device;
+	spin_lock_init(&blkdat->request_queue_lock);
+	atomic_set(&blkdat->requeue_scheduled, 0);
+
+	blkdat->request_queue = blk_init_queue(
+		tapeblock_request_fn,
+		&blkdat->request_queue_lock
+	);
+	if (!blkdat->request_queue)
+		return -ENOMEM;
+
+	elevator_exit(blkdat->request_queue->elevator);
+	rc = elevator_init(blkdat->request_queue, "noop");
+	if (rc)
+		goto cleanup_queue;
+
+	blk_queue_logical_block_size(blkdat->request_queue, TAPEBLOCK_HSEC_SIZE);
+	blk_queue_max_sectors(blkdat->request_queue, TAPEBLOCK_MAX_SEC);
+	blk_queue_max_phys_segments(blkdat->request_queue, -1L);
+	blk_queue_max_hw_segments(blkdat->request_queue, -1L);
+	blk_queue_max_segment_size(blkdat->request_queue, -1L);
+	blk_queue_segment_boundary(blkdat->request_queue, -1L);
+
+	disk = alloc_disk(1);
+	if (!disk) {
+		rc = -ENOMEM;
+		goto cleanup_queue;
+	}
+
+	disk->major = tapeblock_major;
+	disk->first_minor = device->first_minor;
+	disk->fops = &tapeblock_fops;
+	disk->private_data = tape_get_device_reference(device);
+	disk->queue = blkdat->request_queue;
+	set_capacity(disk, 0);
+	sprintf(disk->disk_name, "btibm%d",
+		device->first_minor / TAPE_MINORS_PER_DEV);
+
+	blkdat->disk = disk;
+	blkdat->medium_changed = 1;
+	blkdat->request_queue->queuedata = tape_get_device_reference(device);
+
+	add_disk(disk);
+
+	tape_get_device_reference(device);
+	INIT_WORK(&blkdat->requeue_task, tapeblock_requeue);
+
+	return 0;
+
+cleanup_queue:
+	blk_cleanup_queue(blkdat->request_queue);
+	blkdat->request_queue = NULL;
+
+	return rc;
+}
+
+void
+tapeblock_cleanup_device(struct tape_device *device)
+{
+	flush_scheduled_work();
+	tape_put_device(device);
+
+	if (!device->blk_data.disk) {
+		goto cleanup_queue;
+	}
+
+	del_gendisk(device->blk_data.disk);
+	device->blk_data.disk->private_data =
+		tape_put_device(device->blk_data.disk->private_data);
+	put_disk(device->blk_data.disk);
+
+	device->blk_data.disk = NULL;
+cleanup_queue:
+	device->blk_data.request_queue->queuedata = tape_put_device(device);
+
+	blk_cleanup_queue(device->blk_data.request_queue);
+	device->blk_data.request_queue = NULL;
+}
+
+/*
+ * Detect number of blocks of the tape.
+ * FIXME: can we extent this to detect the blocks size as well ?
+ */
+static int
+tapeblock_revalidate_disk(struct gendisk *disk)
+{
+	struct tape_device *	device;
+	unsigned int		nr_of_blks;
+	int			rc;
+
+	device = (struct tape_device *) disk->private_data;
+	BUG_ON(!device);
+
+	if (!device->blk_data.medium_changed)
+		return 0;
+
+	rc = tape_mtop(device, MTFSFM, 1);
+	if (rc)
+		return rc;
+
+	rc = tape_mtop(device, MTTELL, 1);
+	if (rc < 0)
+		return rc;
+
+	pr_info("%s: Determining the size of the recorded area...\n",
+		dev_name(&device->cdev->dev));
+	DBF_LH(3, "Image file ends at %d\n", rc);
+	nr_of_blks = rc;
+
+	/* This will fail for the first file. Catch the error by checking the
+	 * position. */
+	tape_mtop(device, MTBSF, 1);
+
+	rc = tape_mtop(device, MTTELL, 1);
+	if (rc < 0)
+		return rc;
+
+	if (rc > nr_of_blks)
+		return -EINVAL;
+
+	DBF_LH(3, "Image file starts at %d\n", rc);
+	device->bof = rc;
+	nr_of_blks -= rc;
+
+	pr_info("%s: The size of the recorded area is %i blocks\n",
+		dev_name(&device->cdev->dev), nr_of_blks);
+	set_capacity(device->blk_data.disk,
+		nr_of_blks*(TAPEBLOCK_HSEC_SIZE/512));
+
+	device->blk_data.block_position = 0;
+	device->blk_data.medium_changed = 0;
+	return 0;
+}
+
+static int
+tapeblock_medium_changed(struct gendisk *disk)
+{
+	struct tape_device *device;
+
+	device = (struct tape_device *) disk->private_data;
+	DBF_LH(6, "tapeblock_medium_changed(%p) = %d\n",
+		device, device->blk_data.medium_changed);
+
+	return device->blk_data.medium_changed;
+}
+
+/*
+ * Block frontend tape device open function.
+ */
+static int
+tapeblock_open(struct block_device *bdev, fmode_t mode)
+{
+	struct gendisk *	disk = bdev->bd_disk;
+	struct tape_device *	device;
+	int			rc;
+
+	device = tape_get_device_reference(disk->private_data);
+
+	if (device->required_tapemarks) {
+		DBF_EVENT(2, "TBLOCK: missing tapemarks\n");
+		pr_warning("%s: Opening the tape failed because of missing "
+			   "end-of-file marks\n", dev_name(&device->cdev->dev));
+		rc = -EPERM;
+		goto put_device;
+	}
+
+	rc = tape_open(device);
+	if (rc)
+		goto put_device;
+
+	rc = tapeblock_revalidate_disk(disk);
+	if (rc)
+		goto release;
+
+	/*
+	 * Note: The reference to <device> is hold until the release function
+	 *       is called.
+	 */
+	tape_state_set(device, TS_BLKUSE);
+	return 0;
+
+release:
+	tape_release(device);
+ put_device:
+	tape_put_device(device);
+	return rc;
+}
+
+/*
+ * Block frontend tape device release function.
+ *
+ * Note: One reference to the tape device was made by the open function. So
+ *       we just get the pointer here and release the reference.
+ */
+static int
+tapeblock_release(struct gendisk *disk, fmode_t mode)
+{
+	struct tape_device *device = disk->private_data;
+
+	tape_state_set(device, TS_IN_USE);
+	tape_release(device);
+	tape_put_device(device);
+
+	return 0;
+}
+
+/*
+ * Support of some generic block device IOCTLs.
+ */
+static int
+tapeblock_ioctl(
+	struct block_device *	bdev,
+	fmode_t			mode,
+	unsigned int		command,
+	unsigned long		arg
+) {
+	int rc;
+	int minor;
+	struct gendisk *disk = bdev->bd_disk;
+	struct tape_device *device;
+
+	rc     = 0;
+	BUG_ON(!disk);
+	device = disk->private_data;
+	BUG_ON(!device);
+	minor  = MINOR(bdev->bd_dev);
+
+	DBF_LH(6, "tapeblock_ioctl(0x%0x)\n", command);
+	DBF_LH(6, "device = %d:%d\n", tapeblock_major, minor);
+
+	switch (command) {
+		/* Refuse some IOCTL calls without complaining (mount). */
+		case 0x5310:		/* CDROMMULTISESSION */
+			rc = -EINVAL;
+			break;
+		default:
+			rc = -EINVAL;
+	}
+
+	return rc;
+}
+
+/*
+ * Initialize block device frontend.
+ */
+int
+tapeblock_init(void)
+{
+	int rc;
+
+	/* Register the tape major number to the kernel */
+	rc = register_blkdev(tapeblock_major, "tBLK");
+	if (rc < 0)
+		return rc;
+
+	if (tapeblock_major == 0)
+		tapeblock_major = rc;
+	return 0;
+}
+
+/*
+ * Deregister major for block device frontend
+ */
+void
+tapeblock_exit(void)
+{
+	unregister_blkdev(tapeblock_major, "tBLK");
+}

kernel/drivers/s390/char/tape_char.c

@@ -0,0 +1,495 @@
+/*
+ *  drivers/s390/char/tape_char.c
+ *    character device frontend for tape device driver
+ *
+ *  S390 and zSeries version
+ *    Copyright IBM Corp. 2001,2006
+ *    Author(s): Carsten Otte <cotte@de.ibm.com>
+ *		 Michael Holzheu <holzheu@de.ibm.com>
+ *		 Tuan Ngo-Anh <ngoanh@de.ibm.com>
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/proc_fs.h>
+#include <linux/mtio.h>
+#include <linux/smp_lock.h>
+
+#include <asm/uaccess.h>
+
+#define TAPE_DBF_AREA	tape_core_dbf
+
+#include "tape.h"
+#include "tape_std.h"
+#include "tape_class.h"
+
+#define TAPECHAR_MAJOR		0	/* get dynamic major */
+
+/*
+ * file operation structure for tape character frontend
+ */
+static ssize_t tapechar_read(struct file *, char __user *, size_t, loff_t *);
+static ssize_t tapechar_write(struct file *, const char __user *, size_t, loff_t *);
+static int tapechar_open(struct inode *,struct file *);
+static int tapechar_release(struct inode *,struct file *);
+static int tapechar_ioctl(struct inode *, struct file *, unsigned int,
+			  unsigned long);
+static long tapechar_compat_ioctl(struct file *, unsigned int,
+			  unsigned long);
+
+static const struct file_operations tape_fops =
+{
+	.owner = THIS_MODULE,
+	.read = tapechar_read,
+	.write = tapechar_write,
+	.ioctl = tapechar_ioctl,
+	.compat_ioctl = tapechar_compat_ioctl,
+	.open = tapechar_open,
+	.release = tapechar_release,
+};
+
+static int tapechar_major = TAPECHAR_MAJOR;
+
+/*
+ * This function is called for every new tapedevice
+ */
+int
+tapechar_setup_device(struct tape_device * device)
+{
+	char	device_name[20];
+
+	sprintf(device_name, "ntibm%i", device->first_minor / 2);
+	device->nt = register_tape_dev(
+		&device->cdev->dev,
+		MKDEV(tapechar_major, device->first_minor),
+		&tape_fops,
+		device_name,
+		"non-rewinding"
+	);
+	device_name[0] = 'r';
+	device->rt = register_tape_dev(
+		&device->cdev->dev,
+		MKDEV(tapechar_major, device->first_minor + 1),
+		&tape_fops,
+		device_name,
+		"rewinding"
+	);
+
+	return 0;
+}
+
+void
+tapechar_cleanup_device(struct tape_device *device)
+{
+	unregister_tape_dev(&device->cdev->dev, device->rt);
+	device->rt = NULL;
+	unregister_tape_dev(&device->cdev->dev, device->nt);
+	device->nt = NULL;
+}
+
+static int
+tapechar_check_idalbuffer(struct tape_device *device, size_t block_size)
+{
+	struct idal_buffer *new;
+
+	if (device->char_data.idal_buf != NULL &&
+	    device->char_data.idal_buf->size == block_size)
+		return 0;
+
+	if (block_size > MAX_BLOCKSIZE) {
+		DBF_EVENT(3, "Invalid blocksize (%zd > %d)\n",
+			block_size, MAX_BLOCKSIZE);
+		return -EINVAL;
+	}
+
+	/* The current idal buffer is not correct. Allocate a new one. */
+	new = idal_buffer_alloc(block_size, 0);
+	if (IS_ERR(new))
+		return -ENOMEM;
+
+	if (device->char_data.idal_buf != NULL)
+		idal_buffer_free(device->char_data.idal_buf);
+
+	device->char_data.idal_buf = new;
+
+	return 0;
+}
+
+/*
+ * Tape device read function
+ */
+static ssize_t
+tapechar_read(struct file *filp, char __user *data, size_t count, loff_t *ppos)
+{
+	struct tape_device *device;
+	struct tape_request *request;
+	size_t block_size;
+	int rc;
+
+	DBF_EVENT(6, "TCHAR:read\n");
+	device = (struct tape_device *) filp->private_data;
+
+	/*
+	 * If the tape isn't terminated yet, do it now. And since we then
+	 * are at the end of the tape there wouldn't be anything to read
+	 * anyways. So we return immediatly.
+	 */
+	if(device->required_tapemarks) {
+		return tape_std_terminate_write(device);
+	}
+
+	/* Find out block size to use */
+	if (device->char_data.block_size != 0) {
+		if (count < device->char_data.block_size) {
+			DBF_EVENT(3, "TCHAR:read smaller than block "
+				  "size was requested\n");
+			return -EINVAL;
+		}
+		block_size = device->char_data.block_size;
+	} else {
+		block_size = count;
+	}
+
+	rc = tapechar_check_idalbuffer(device, block_size);
+	if (rc)
+		return rc;
+
+#ifdef CONFIG_S390_TAPE_BLOCK
+	/* Changes position. */
+	device->blk_data.medium_changed = 1;
+#endif
+
+	DBF_EVENT(6, "TCHAR:nbytes: %lx\n", block_size);
+	/* Let the discipline build the ccw chain. */
+	request = device->discipline->read_block(device, block_size);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	/* Execute it. */
+	rc = tape_do_io(device, request);
+	if (rc == 0) {
+		rc = block_size - request->rescnt;
+		DBF_EVENT(6, "TCHAR:rbytes:  %x\n", rc);
+		filp->f_pos += rc;
+		/* Copy data from idal buffer to user space. */
+		if (idal_buffer_to_user(device->char_data.idal_buf,
+					data, rc) != 0)
+			rc = -EFAULT;
+	}
+	tape_free_request(request);
+	return rc;
+}
+
+/*
+ * Tape device write function
+ */
+static ssize_t
+tapechar_write(struct file *filp, const char __user *data, size_t count, loff_t *ppos)
+{
+	struct tape_device *device;
+	struct tape_request *request;
+	size_t block_size;
+	size_t written;
+	int nblocks;
+	int i, rc;
+
+	DBF_EVENT(6, "TCHAR:write\n");
+	device = (struct tape_device *) filp->private_data;
+	/* Find out block size and number of blocks */
+	if (device->char_data.block_size != 0) {
+		if (count < device->char_data.block_size) {
+			DBF_EVENT(3, "TCHAR:write smaller than block "
+				  "size was requested\n");
+			return -EINVAL;
+		}
+		block_size = device->char_data.block_size;
+		nblocks = count / block_size;
+	} else {
+		block_size = count;
+		nblocks = 1;
+	}
+
+	rc = tapechar_check_idalbuffer(device, block_size);
+	if (rc)
+		return rc;
+
+#ifdef CONFIG_S390_TAPE_BLOCK
+	/* Changes position. */
+	device->blk_data.medium_changed = 1;
+#endif
+
+	DBF_EVENT(6,"TCHAR:nbytes: %lx\n", block_size);
+	DBF_EVENT(6, "TCHAR:nblocks: %x\n", nblocks);
+	/* Let the discipline build the ccw chain. */
+	request = device->discipline->write_block(device, block_size);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	rc = 0;
+	written = 0;
+	for (i = 0; i < nblocks; i++) {
+		/* Copy data from user space to idal buffer. */
+		if (idal_buffer_from_user(device->char_data.idal_buf,
+					  data, block_size)) {
+			rc = -EFAULT;
+			break;
+		}
+		rc = tape_do_io(device, request);
+		if (rc)
+			break;
+		DBF_EVENT(6, "TCHAR:wbytes: %lx\n",
+			  block_size - request->rescnt);
+		filp->f_pos += block_size - request->rescnt;
+		written += block_size - request->rescnt;
+		if (request->rescnt != 0)
+			break;
+		data += block_size;
+	}
+	tape_free_request(request);
+	if (rc == -ENOSPC) {
+		/*
+		 * Ok, the device has no more space. It has NOT written
+		 * the block.
+		 */
+		if (device->discipline->process_eov)
+			device->discipline->process_eov(device);
+		if (written > 0)
+			rc = 0;
+
+	}
+
+	/*
+	 * After doing a write we always need two tapemarks to correctly
+	 * terminate the tape (one to terminate the file, the second to
+	 * flag the end of recorded data.
+	 * Since process_eov positions the tape in front of the written
+	 * tapemark it doesn't hurt to write two marks again.
+	 */
+	if (!rc)
+		device->required_tapemarks = 2;
+
+	return rc ? rc : written;
+}
+
+/*
+ * Character frontend tape device open function.
+ */
+static int
+tapechar_open (struct inode *inode, struct file *filp)
+{
+	struct tape_device *device;
+	int minor, rc;
+
+	DBF_EVENT(6, "TCHAR:open: %i:%i\n",
+		imajor(filp->f_path.dentry->d_inode),
+		iminor(filp->f_path.dentry->d_inode));
+
+	if (imajor(filp->f_path.dentry->d_inode) != tapechar_major)
+		return -ENODEV;
+
+	lock_kernel();
+	minor = iminor(filp->f_path.dentry->d_inode);
+	device = tape_get_device(minor / TAPE_MINORS_PER_DEV);
+	if (IS_ERR(device)) {
+		DBF_EVENT(3, "TCHAR:open: tape_get_device() failed\n");
+		rc = PTR_ERR(device);
+		goto out;
+	}
+
+
+	rc = tape_open(device);
+	if (rc == 0) {
+		filp->private_data = device;
+		rc = nonseekable_open(inode, filp);
+	}
+	else
+		tape_put_device(device);
+
+out:
+	unlock_kernel();
+	return rc;
+}
+
+/*
+ * Character frontend tape device release function.
+ */
+
+static int
+tapechar_release(struct inode *inode, struct file *filp)
+{
+	struct tape_device *device;
+
+	DBF_EVENT(6, "TCHAR:release: %x\n", iminor(inode));
+	device = (struct tape_device *) filp->private_data;
+
+	/*
+	 * If this is the rewinding tape minor then rewind. In that case we
+	 * write all required tapemarks. Otherwise only one to terminate the
+	 * file.
+	 */
+	if ((iminor(inode) & 1) != 0) {
+		if (device->required_tapemarks)
+			tape_std_terminate_write(device);
+		tape_mtop(device, MTREW, 1);
+	} else {
+		if (device->required_tapemarks > 1) {
+			if (tape_mtop(device, MTWEOF, 1) == 0)
+				device->required_tapemarks--;
+		}
+	}
+
+	if (device->char_data.idal_buf != NULL) {
+		idal_buffer_free(device->char_data.idal_buf);
+		device->char_data.idal_buf = NULL;
+	}
+	tape_release(device);
+	filp->private_data = tape_put_device(device);
+
+	return 0;
+}
+
+/*
+ * Tape device io controls.
+ */
+static int
+tapechar_ioctl(struct inode *inp, struct file *filp,
+	       unsigned int no, unsigned long data)
+{
+	struct tape_device *device;
+	int rc;
+
+	DBF_EVENT(6, "TCHAR:ioct\n");
+
+	device = (struct tape_device *) filp->private_data;
+
+	if (no == MTIOCTOP) {
+		struct mtop op;
+
+		if (copy_from_user(&op, (char __user *) data, sizeof(op)) != 0)
+			return -EFAULT;
+		if (op.mt_count < 0)
+			return -EINVAL;
+
+		/*
+		 * Operations that change tape position should write final
+		 * tapemarks.
+		 */
+		switch (op.mt_op) {
+			case MTFSF:
+			case MTBSF:
+			case MTFSR:
+			case MTBSR:
+			case MTREW:
+			case MTOFFL:
+			case MTEOM:
+			case MTRETEN:
+			case MTBSFM:
+			case MTFSFM:
+			case MTSEEK:
+#ifdef CONFIG_S390_TAPE_BLOCK
+				device->blk_data.medium_changed = 1;
+#endif
+				if (device->required_tapemarks)
+					tape_std_terminate_write(device);
+			default:
+				;
+		}
+		rc = tape_mtop(device, op.mt_op, op.mt_count);
+
+		if (op.mt_op == MTWEOF && rc == 0) {
+			if (op.mt_count > device->required_tapemarks)
+				device->required_tapemarks = 0;
+			else
+				device->required_tapemarks -= op.mt_count;
+		}
+		return rc;
+	}
+	if (no == MTIOCPOS) {
+		/* MTIOCPOS: query the tape position. */
+		struct mtpos pos;
+
+		rc = tape_mtop(device, MTTELL, 1);
+		if (rc < 0)
+			return rc;
+		pos.mt_blkno = rc;
+		if (copy_to_user((char __user *) data, &pos, sizeof(pos)) != 0)
+			return -EFAULT;
+		return 0;
+	}
+	if (no == MTIOCGET) {
+		/* MTIOCGET: query the tape drive status. */
+		struct mtget get;
+
+		memset(&get, 0, sizeof(get));
+		get.mt_type = MT_ISUNKNOWN;
+		get.mt_resid = 0 /* device->devstat.rescnt */;
+		get.mt_dsreg = device->tape_state;
+		/* FIXME: mt_gstat, mt_erreg, mt_fileno */
+		get.mt_gstat = 0;
+		get.mt_erreg = 0;
+		get.mt_fileno = 0;
+		get.mt_gstat  = device->tape_generic_status;
+
+		if (device->medium_state == MS_LOADED) {
+			rc = tape_mtop(device, MTTELL, 1);
+
+			if (rc < 0)
+				return rc;
+
+			if (rc == 0)
+				get.mt_gstat |= GMT_BOT(~0);
+
+			get.mt_blkno = rc;
+		}
+
+		if (copy_to_user((char __user *) data, &get, sizeof(get)) != 0)
+			return -EFAULT;
+
+		return 0;
+	}
+	/* Try the discipline ioctl function. */
+	if (device->discipline->ioctl_fn == NULL)
+		return -EINVAL;
+	return device->discipline->ioctl_fn(device, no, data);
+}
+
+static long
+tapechar_compat_ioctl(struct file *filp, unsigned int no, unsigned long data)
+{
+	struct tape_device *device = filp->private_data;
+	int rval = -ENOIOCTLCMD;
+
+	if (device->discipline->ioctl_fn) {
+		lock_kernel();
+		rval = device->discipline->ioctl_fn(device, no, data);
+		unlock_kernel();
+		if (rval == -EINVAL)
+			rval = -ENOIOCTLCMD;
+	}
+
+	return rval;
+}
+
+/*
+ * Initialize character device frontend.
+ */
+int
+tapechar_init (void)
+{
+	dev_t	dev;
+
+	if (alloc_chrdev_region(&dev, 0, 256, "tape") != 0)
+		return -1;
+
+	tapechar_major = MAJOR(dev);
+
+	return 0;
+}
+
+/*
+ * cleanup
+ */
+void
+tapechar_exit(void)
+{
+	unregister_chrdev_region(MKDEV(tapechar_major, 0), 256);
+}

kernel/drivers/s390/char/tape_class.c

@@ -0,0 +1,127 @@
+/*
+ * (C) Copyright IBM Corp. 2004
+ * tape_class.c
+ *
+ * Tape class device support
+ *
+ * Author: Stefan Bader <shbader@de.ibm.com>
+ * Based on simple class device code by Greg K-H
+ */
+#include "tape_class.h"
+
+MODULE_AUTHOR("Stefan Bader <shbader@de.ibm.com>");
+MODULE_DESCRIPTION(
+	"(C) Copyright IBM Corp. 2004   All Rights Reserved.\n"
+	"tape_class.c"
+);
+MODULE_LICENSE("GPL");
+
+static struct class *tape_class;
+
+/*
+ * Register a tape device and return a pointer to the cdev structure.
+ *
+ * device
+ *	The pointer to the struct device of the physical (base) device.
+ * drivername
+ *	The pointer to the drivers name for it's character devices.
+ * dev
+ *	The intended major/minor number. The major number may be 0 to
+ *	get a dynamic major number.
+ * fops
+ *	The pointer to the drivers file operations for the tape device.
+ * devname
+ *	The pointer to the name of the character device.
+ */
+struct tape_class_device *register_tape_dev(
+	struct device *		device,
+	dev_t			dev,
+	const struct file_operations *fops,
+	char *			device_name,
+	char *			mode_name)
+{
+	struct tape_class_device *	tcd;
+	int		rc;
+	char *		s;
+
+	tcd = kzalloc(sizeof(struct tape_class_device), GFP_KERNEL);
+	if (!tcd)
+		return ERR_PTR(-ENOMEM);
+
+	strncpy(tcd->device_name, device_name, TAPECLASS_NAME_LEN);
+	for (s = strchr(tcd->device_name, '/'); s; s = strchr(s, '/'))
+		*s = '!';
+	strncpy(tcd->mode_name, mode_name, TAPECLASS_NAME_LEN);
+	for (s = strchr(tcd->mode_name, '/'); s; s = strchr(s, '/'))
+		*s = '!';
+
+	tcd->char_device = cdev_alloc();
+	if (!tcd->char_device) {
+		rc = -ENOMEM;
+		goto fail_with_tcd;
+	}
+
+	tcd->char_device->owner = fops->owner;
+	tcd->char_device->ops   = fops;
+	tcd->char_device->dev   = dev;
+
+	rc = cdev_add(tcd->char_device, tcd->char_device->dev, 1);
+	if (rc)
+		goto fail_with_cdev;
+
+	tcd->class_device = device_create(tape_class, device,
+					  tcd->char_device->dev, NULL,
+					  "%s", tcd->device_name);
+	rc = IS_ERR(tcd->class_device) ? PTR_ERR(tcd->class_device) : 0;
+	if (rc)
+		goto fail_with_cdev;
+	rc = sysfs_create_link(
+		&device->kobj,
+		&tcd->class_device->kobj,
+		tcd->mode_name
+	);
+	if (rc)
+		goto fail_with_class_device;
+
+	return tcd;
+
+fail_with_class_device:
+	device_destroy(tape_class, tcd->char_device->dev);
+
+fail_with_cdev:
+	cdev_del(tcd->char_device);
+
+fail_with_tcd:
+	kfree(tcd);
+
+	return ERR_PTR(rc);
+}
+EXPORT_SYMBOL(register_tape_dev);
+
+void unregister_tape_dev(struct device *device, struct tape_class_device *tcd)
+{
+	if (tcd != NULL && !IS_ERR(tcd)) {
+		sysfs_remove_link(&device->kobj, tcd->mode_name);
+		device_destroy(tape_class, tcd->char_device->dev);
+		cdev_del(tcd->char_device);
+		kfree(tcd);
+	}
+}
+EXPORT_SYMBOL(unregister_tape_dev);
+
+
+static int __init tape_init(void)
+{
+	tape_class = class_create(THIS_MODULE, "tape390");
+
+	return 0;
+}
+
+static void __exit tape_exit(void)
+{
+	class_destroy(tape_class);
+	tape_class = NULL;
+}
+
+postcore_initcall(tape_init);
+module_exit(tape_exit);

kernel/drivers/s390/char/tape_class.h

@@ -0,0 +1,61 @@
+/*
+ * (C) Copyright IBM Corp. 2004   All Rights Reserved.
+ * tape_class.h
+ *
+ * Tape class device support
+ *
+ * Author: Stefan Bader <shbader@de.ibm.com>
+ * Based on simple class device code by Greg K-H
+ */
+#ifndef __TAPE_CLASS_H__
+#define __TAPE_CLASS_H__
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/major.h>
+#include <linux/kobject.h>
+#include <linux/kobj_map.h>
+#include <linux/cdev.h>
+
+#include <linux/device.h>
+#include <linux/kdev_t.h>
+
+#define TAPECLASS_NAME_LEN	32
+
+struct tape_class_device {
+	struct cdev		*char_device;
+	struct device		*class_device;
+	char			device_name[TAPECLASS_NAME_LEN];
+	char			mode_name[TAPECLASS_NAME_LEN];
+};
+
+/*
+ * Register a tape device and return a pointer to the tape class device
+ * created by the call.
+ *
+ * device
+ *	The pointer to the struct device of the physical (base) device.
+ * dev
+ *	The intended major/minor number. The major number may be 0 to
+ *	get a dynamic major number.
+ * fops
+ *	The pointer to the drivers file operations for the tape device.
+ * device_name
+ *	Pointer to the logical device name (will also be used as kobject name
+ *	of the cdev). This can also be called the name of the tape class
+ *	device.
+ * mode_name
+ *	Points to the name of the tape mode. This creates a link with that
+ *	name from the physical device to the logical device (class).
+ */
+struct tape_class_device *register_tape_dev(
+	struct device *		device,
+	dev_t			dev,
+	const struct file_operations *fops,
+	char *			device_name,
+	char *			node_name
+);
+void unregister_tape_dev(struct device *device, struct tape_class_device *tcd);
+
+#endif /* __TAPE_CLASS_H__ */

kernel/drivers/s390/char/tape_proc.c

@@ -0,0 +1,141 @@
+/*
+ *  drivers/s390/char/tape.c
+ *    tape device driver for S/390 and zSeries tapes.
+ *
+ *  S390 and zSeries version
+ *    Copyright (C) 2001 IBM Corporation
+ *    Author(s): Carsten Otte <cotte@de.ibm.com>
+ *		 Michael Holzheu <holzheu@de.ibm.com>
+ *		 Tuan Ngo-Anh <ngoanh@de.ibm.com>
+ *
+ * PROCFS Functions
+ */
+
+#include <linux/module.h>
+#include <linux/vmalloc.h>
+#include <linux/seq_file.h>
+#include <linux/proc_fs.h>
+
+#define TAPE_DBF_AREA	tape_core_dbf
+
+#include "tape.h"
+
+static const char *tape_med_st_verbose[MS_SIZE] =
+{
+	[MS_UNKNOWN] = "UNKNOWN ",
+	[MS_LOADED] = "LOADED  ",
+	[MS_UNLOADED] = "UNLOADED"
+};
+
+/* our proc tapedevices entry */
+static struct proc_dir_entry *tape_proc_devices;
+
+/*
+ * Show function for /proc/tapedevices
+ */
+static int tape_proc_show(struct seq_file *m, void *v)
+{
+	struct tape_device *device;
+	struct tape_request *request;
+	const char *str;
+	unsigned long n;
+
+	n = (unsigned long) v - 1;
+	if (!n) {
+		seq_printf(m, "TapeNo\tBusID      CuType/Model\t"
+			"DevType/Model\tBlkSize\tState\tOp\tMedState\n");
+	}
+	device = tape_get_device(n);
+	if (IS_ERR(device))
+		return 0;
+	spin_lock_irq(get_ccwdev_lock(device->cdev));
+	seq_printf(m, "%d\t", (int) n);
+	seq_printf(m, "%-10.10s ", dev_name(&device->cdev->dev));
+	seq_printf(m, "%04X/", device->cdev->id.cu_type);
+	seq_printf(m, "%02X\t", device->cdev->id.cu_model);
+	seq_printf(m, "%04X/", device->cdev->id.dev_type);
+	seq_printf(m, "%02X\t\t", device->cdev->id.dev_model);
+	if (device->char_data.block_size == 0)
+		seq_printf(m, "auto\t");
+	else
+		seq_printf(m, "%i\t", device->char_data.block_size);
+	if (device->tape_state >= 0 &&
+	    device->tape_state < TS_SIZE)
+		str = tape_state_verbose[device->tape_state];
+	else
+		str = "UNKNOWN";
+	seq_printf(m, "%s\t", str);
+	if (!list_empty(&device->req_queue)) {
+		request = list_entry(device->req_queue.next,
+				     struct tape_request, list);
+		str = tape_op_verbose[request->op];
+	} else
+		str = "---";
+	seq_printf(m, "%s\t", str);
+	seq_printf(m, "%s\n", tape_med_st_verbose[device->medium_state]);
+	spin_unlock_irq(get_ccwdev_lock(device->cdev));
+	tape_put_device(device);
+        return 0;
+}
+
+static void *tape_proc_start(struct seq_file *m, loff_t *pos)
+{
+	if (*pos >= 256 / TAPE_MINORS_PER_DEV)
+		return NULL;
+	return (void *)((unsigned long) *pos + 1);
+}
+
+static void *tape_proc_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return tape_proc_start(m, pos);
+}
+
+static void tape_proc_stop(struct seq_file *m, void *v)
+{
+}
+
+static const struct seq_operations tape_proc_seq = {
+	.start		= tape_proc_start,
+	.next		= tape_proc_next,
+	.stop		= tape_proc_stop,
+	.show		= tape_proc_show,
+};
+
+static int tape_proc_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &tape_proc_seq);
+}
+
+static const struct file_operations tape_proc_ops =
+{
+	.owner		= THIS_MODULE,
+	.open		= tape_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+/*
+ * Initialize procfs stuff on startup
+ */
+void
+tape_proc_init(void)
+{
+	tape_proc_devices =
+		proc_create("tapedevices", S_IFREG | S_IRUGO | S_IWUSR, NULL,
+			    &tape_proc_ops);
+	if (tape_proc_devices == NULL) {
+		return;
+	}
+}
+
+/*
+ * Cleanup all stuff registered to the procfs
+ */
+void
+tape_proc_cleanup(void)
+{
+	if (tape_proc_devices != NULL)
+		remove_proc_entry ("tapedevices", NULL);
+}

kernel/drivers/s390/char/tape_std.c

@@ -0,0 +1,748 @@
+/*
+ *  drivers/s390/char/tape_std.c
+ *    standard tape device functions for ibm tapes.
+ *
+ *  S390 and zSeries version
+ *    Copyright (C) 2001,2002 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Author(s): Carsten Otte <cotte@de.ibm.com>
+ *		 Michael Holzheu <holzheu@de.ibm.com>
+ *		 Tuan Ngo-Anh <ngoanh@de.ibm.com>
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>
+ *		 Stefan Bader <shbader@de.ibm.com>
+ */
+
+#include <linux/stddef.h>
+#include <linux/kernel.h>
+#include <linux/bio.h>
+#include <linux/timer.h>
+
+#include <asm/types.h>
+#include <asm/idals.h>
+#include <asm/ebcdic.h>
+#include <asm/tape390.h>
+
+#define TAPE_DBF_AREA	tape_core_dbf
+
+#include "tape.h"
+#include "tape_std.h"
+
+/*
+ * tape_std_assign
+ */
+static void
+tape_std_assign_timeout(unsigned long data)
+{
+	struct tape_request *	request;
+	struct tape_device *	device;
+	int rc;
+
+	request = (struct tape_request *) data;
+	device = request->device;
+	BUG_ON(!device);
+
+	DBF_EVENT(3, "%08x: Assignment timeout. Device busy.\n",
+			device->cdev_id);
+	rc = tape_cancel_io(device, request);
+	if(rc)
+		DBF_EVENT(3, "(%s): Assign timeout: Cancel failed with rc = %i\n",
+			dev_name(&device->cdev->dev), rc);
+}
+
+int
+tape_std_assign(struct tape_device *device)
+{
+	int                  rc;
+	struct timer_list    timeout;
+	struct tape_request *request;
+
+	request = tape_alloc_request(2, 11);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+
+	request->op = TO_ASSIGN;
+	tape_ccw_cc(request->cpaddr, ASSIGN, 11, request->cpdata);
+	tape_ccw_end(request->cpaddr + 1, NOP, 0, NULL);
+
+	/*
+	 * The assign command sometimes blocks if the device is assigned
+	 * to another host (actually this shouldn't happen but it does).
+	 * So we set up a timeout for this call.
+	 */
+	init_timer_on_stack(&timeout);
+	timeout.function = tape_std_assign_timeout;
+	timeout.data     = (unsigned long) request;
+	timeout.expires  = jiffies + 2 * HZ;
+	add_timer(&timeout);
+
+	rc = tape_do_io_interruptible(device, request);
+
+	del_timer(&timeout);
+
+	if (rc != 0) {
+		DBF_EVENT(3, "%08x: assign failed - device might be busy\n",
+			device->cdev_id);
+	} else {
+		DBF_EVENT(3, "%08x: Tape assigned\n", device->cdev_id);
+	}
+	tape_free_request(request);
+	return rc;
+}
+
+/*
+ * tape_std_unassign
+ */
+int
+tape_std_unassign (struct tape_device *device)
+{
+	int                  rc;
+	struct tape_request *request;
+
+	if (device->tape_state == TS_NOT_OPER) {
+		DBF_EVENT(3, "(%08x): Can't unassign device\n",
+			device->cdev_id);
+		return -EIO;
+	}
+
+	request = tape_alloc_request(2, 11);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+
+	request->op = TO_UNASSIGN;
+	tape_ccw_cc(request->cpaddr, UNASSIGN, 11, request->cpdata);
+	tape_ccw_end(request->cpaddr + 1, NOP, 0, NULL);
+
+	if ((rc = tape_do_io(device, request)) != 0) {
+		DBF_EVENT(3, "%08x: Unassign failed\n", device->cdev_id);
+	} else {
+		DBF_EVENT(3, "%08x: Tape unassigned\n", device->cdev_id);
+	}
+	tape_free_request(request);
+	return rc;
+}
+
+/*
+ * TAPE390_DISPLAY: Show a string on the tape display.
+ */
+int
+tape_std_display(struct tape_device *device, struct display_struct *disp)
+{
+	struct tape_request *request;
+	int rc;
+
+	request = tape_alloc_request(2, 17);
+	if (IS_ERR(request)) {
+		DBF_EVENT(3, "TAPE: load display failed\n");
+		return PTR_ERR(request);
+	}
+	request->op = TO_DIS;
+
+	*(unsigned char *) request->cpdata = disp->cntrl;
+	DBF_EVENT(5, "TAPE: display cntrl=%04x\n", disp->cntrl);
+	memcpy(((unsigned char *) request->cpdata) + 1, disp->message1, 8);
+	memcpy(((unsigned char *) request->cpdata) + 9, disp->message2, 8);
+	ASCEBC(((unsigned char*) request->cpdata) + 1, 16);
+
+	tape_ccw_cc(request->cpaddr, LOAD_DISPLAY, 17, request->cpdata);
+	tape_ccw_end(request->cpaddr + 1, NOP, 0, NULL);
+
+	rc = tape_do_io_interruptible(device, request);
+	tape_free_request(request);
+	return rc;
+}
+
+/*
+ * Read block id.
+ */
+int
+tape_std_read_block_id(struct tape_device *device, __u64 *id)
+{
+	struct tape_request *request;
+	int rc;
+
+	request = tape_alloc_request(3, 8);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_RBI;
+	/* setup ccws */
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_cc(request->cpaddr + 1, READ_BLOCK_ID, 8, request->cpdata);
+	tape_ccw_end(request->cpaddr + 2, NOP, 0, NULL);
+	/* execute it */
+	rc = tape_do_io(device, request);
+	if (rc == 0)
+		/* Get result from read buffer. */
+		*id = *(__u64 *) request->cpdata;
+	tape_free_request(request);
+	return rc;
+}
+
+int
+tape_std_terminate_write(struct tape_device *device)
+{
+	int rc;
+
+	if(device->required_tapemarks == 0)
+		return 0;
+
+	DBF_LH(5, "tape%d: terminate write %dxEOF\n", device->first_minor,
+		device->required_tapemarks);
+
+	rc = tape_mtop(device, MTWEOF, device->required_tapemarks);
+	if (rc)
+		return rc;
+
+	device->required_tapemarks = 0;
+	return tape_mtop(device, MTBSR, 1);
+}
+
+/*
+ * MTLOAD: Loads the tape.
+ * The default implementation just wait until the tape medium state changes
+ * to MS_LOADED.
+ */
+int
+tape_std_mtload(struct tape_device *device, int count)
+{
+	return wait_event_interruptible(device->state_change_wq,
+		(device->medium_state == MS_LOADED));
+}
+
+/*
+ * MTSETBLK: Set block size.
+ */
+int
+tape_std_mtsetblk(struct tape_device *device, int count)
+{
+	struct idal_buffer *new;
+
+	DBF_LH(6, "tape_std_mtsetblk(%d)\n", count);
+	if (count <= 0) {
+		/*
+		 * Just set block_size to 0. tapechar_read/tapechar_write
+		 * will realloc the idal buffer if a bigger one than the
+		 * current is needed.
+		 */
+		device->char_data.block_size = 0;
+		return 0;
+	}
+	if (device->char_data.idal_buf != NULL &&
+	    device->char_data.idal_buf->size == count)
+		/* We already have a idal buffer of that size. */
+		return 0;
+
+	if (count > MAX_BLOCKSIZE) {
+		DBF_EVENT(3, "Invalid block size (%d > %d) given.\n",
+			count, MAX_BLOCKSIZE);
+		return -EINVAL;
+	}
+
+	/* Allocate a new idal buffer. */
+	new = idal_buffer_alloc(count, 0);
+	if (IS_ERR(new))
+		return -ENOMEM;
+	if (device->char_data.idal_buf != NULL)
+		idal_buffer_free(device->char_data.idal_buf);
+	device->char_data.idal_buf = new;
+	device->char_data.block_size = count;
+
+	DBF_LH(6, "new blocksize is %d\n", device->char_data.block_size);
+
+	return 0;
+}
+
+/*
+ * MTRESET: Set block size to 0.
+ */
+int
+tape_std_mtreset(struct tape_device *device, int count)
+{
+	DBF_EVENT(6, "TCHAR:devreset:\n");
+	device->char_data.block_size = 0;
+	return 0;
+}
+
+/*
+ * MTFSF: Forward space over 'count' file marks. The tape is positioned
+ * at the EOT (End of Tape) side of the file mark.
+ */
+int
+tape_std_mtfsf(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+	struct ccw1 *ccw;
+
+	request = tape_alloc_request(mt_count + 2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_FSF;
+	/* setup ccws */
+	ccw = tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1,
+			  device->modeset_byte);
+	ccw = tape_ccw_repeat(ccw, FORSPACEFILE, mt_count);
+	ccw = tape_ccw_end(ccw, NOP, 0, NULL);
+
+	/* execute it */
+	return tape_do_io_free(device, request);
+}
+
+/*
+ * MTFSR: Forward space over 'count' tape blocks (blocksize is set
+ * via MTSETBLK.
+ */
+int
+tape_std_mtfsr(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+	struct ccw1 *ccw;
+	int rc;
+
+	request = tape_alloc_request(mt_count + 2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_FSB;
+	/* setup ccws */
+	ccw = tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1,
+			  device->modeset_byte);
+	ccw = tape_ccw_repeat(ccw, FORSPACEBLOCK, mt_count);
+	ccw = tape_ccw_end(ccw, NOP, 0, NULL);
+
+	/* execute it */
+	rc = tape_do_io(device, request);
+	if (rc == 0 && request->rescnt > 0) {
+		DBF_LH(3, "FSR over tapemark\n");
+		rc = 1;
+	}
+	tape_free_request(request);
+
+	return rc;
+}
+
+/*
+ * MTBSR: Backward space over 'count' tape blocks.
+ * (blocksize is set via MTSETBLK.
+ */
+int
+tape_std_mtbsr(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+	struct ccw1 *ccw;
+	int rc;
+
+	request = tape_alloc_request(mt_count + 2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_BSB;
+	/* setup ccws */
+	ccw = tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1,
+			  device->modeset_byte);
+	ccw = tape_ccw_repeat(ccw, BACKSPACEBLOCK, mt_count);
+	ccw = tape_ccw_end(ccw, NOP, 0, NULL);
+
+	/* execute it */
+	rc = tape_do_io(device, request);
+	if (rc == 0 && request->rescnt > 0) {
+		DBF_LH(3, "BSR over tapemark\n");
+		rc = 1;
+	}
+	tape_free_request(request);
+
+	return rc;
+}
+
+/*
+ * MTWEOF: Write 'count' file marks at the current position.
+ */
+int
+tape_std_mtweof(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+	struct ccw1 *ccw;
+
+	request = tape_alloc_request(mt_count + 2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_WTM;
+	/* setup ccws */
+	ccw = tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1,
+			  device->modeset_byte);
+	ccw = tape_ccw_repeat(ccw, WRITETAPEMARK, mt_count);
+	ccw = tape_ccw_end(ccw, NOP, 0, NULL);
+
+	/* execute it */
+	return tape_do_io_free(device, request);
+}
+
+/*
+ * MTBSFM: Backward space over 'count' file marks.
+ * The tape is positioned at the BOT (Begin Of Tape) side of the
+ * last skipped file mark.
+ */
+int
+tape_std_mtbsfm(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+	struct ccw1 *ccw;
+
+	request = tape_alloc_request(mt_count + 2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_BSF;
+	/* setup ccws */
+	ccw = tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1,
+			  device->modeset_byte);
+	ccw = tape_ccw_repeat(ccw, BACKSPACEFILE, mt_count);
+	ccw = tape_ccw_end(ccw, NOP, 0, NULL);
+
+	/* execute it */
+	return tape_do_io_free(device, request);
+}
+
+/*
+ * MTBSF: Backward space over 'count' file marks. The tape is positioned at
+ * the EOT (End of Tape) side of the last skipped file mark.
+ */
+int
+tape_std_mtbsf(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+	struct ccw1 *ccw;
+	int rc;
+
+	request = tape_alloc_request(mt_count + 2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_BSF;
+	/* setup ccws */
+	ccw = tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1,
+			  device->modeset_byte);
+	ccw = tape_ccw_repeat(ccw, BACKSPACEFILE, mt_count);
+	ccw = tape_ccw_end(ccw, NOP, 0, NULL);
+	/* execute it */
+	rc = tape_do_io_free(device, request);
+	if (rc == 0) {
+		rc = tape_mtop(device, MTFSR, 1);
+		if (rc > 0)
+			rc = 0;
+	}
+	return rc;
+}
+
+/*
+ * MTFSFM: Forward space over 'count' file marks.
+ * The tape is positioned at the BOT (Begin Of Tape) side
+ * of the last skipped file mark.
+ */
+int
+tape_std_mtfsfm(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+	struct ccw1 *ccw;
+	int rc;
+
+	request = tape_alloc_request(mt_count + 2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_FSF;
+	/* setup ccws */
+	ccw = tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1,
+			  device->modeset_byte);
+	ccw = tape_ccw_repeat(ccw, FORSPACEFILE, mt_count);
+	ccw = tape_ccw_end(ccw, NOP, 0, NULL);
+	/* execute it */
+	rc = tape_do_io_free(device, request);
+	if (rc == 0) {
+		rc = tape_mtop(device, MTBSR, 1);
+		if (rc > 0)
+			rc = 0;
+	}
+
+	return rc;
+}
+
+/*
+ * MTREW: Rewind the tape.
+ */
+int
+tape_std_mtrew(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+
+	request = tape_alloc_request(3, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_REW;
+	/* setup ccws */
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1,
+		    device->modeset_byte);
+	tape_ccw_cc(request->cpaddr + 1, REWIND, 0, NULL);
+	tape_ccw_end(request->cpaddr + 2, NOP, 0, NULL);
+
+	/* execute it */
+	return tape_do_io_free(device, request);
+}
+
+/*
+ * MTOFFL: Rewind the tape and put the drive off-line.
+ * Implement 'rewind unload'
+ */
+int
+tape_std_mtoffl(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+
+	request = tape_alloc_request(3, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_RUN;
+	/* setup ccws */
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_cc(request->cpaddr + 1, REWIND_UNLOAD, 0, NULL);
+	tape_ccw_end(request->cpaddr + 2, NOP, 0, NULL);
+
+	/* execute it */
+	return tape_do_io_free(device, request);
+}
+
+/*
+ * MTNOP: 'No operation'.
+ */
+int
+tape_std_mtnop(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+
+	request = tape_alloc_request(2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_NOP;
+	/* setup ccws */
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_end(request->cpaddr + 1, NOP, 0, NULL);
+	/* execute it */
+	return tape_do_io_free(device, request);
+}
+
+/*
+ * MTEOM: positions at the end of the portion of the tape already used
+ * for recordind data. MTEOM positions after the last file mark, ready for
+ * appending another file.
+ */
+int
+tape_std_mteom(struct tape_device *device, int mt_count)
+{
+	int rc;
+
+	/*
+	 * Seek from the beginning of tape (rewind).
+	 */
+	if ((rc = tape_mtop(device, MTREW, 1)) < 0)
+		return rc;
+
+	/*
+	 * The logical end of volume is given by two sewuential tapemarks.
+	 * Look for this by skipping to the next file (over one tapemark)
+	 * and then test for another one (fsr returns 1 if a tapemark was
+	 * encountered).
+	 */
+	do {
+		if ((rc = tape_mtop(device, MTFSF, 1)) < 0)
+			return rc;
+		if ((rc = tape_mtop(device, MTFSR, 1)) < 0)
+			return rc;
+	} while (rc == 0);
+
+	return tape_mtop(device, MTBSR, 1);
+}
+
+/*
+ * MTRETEN: Retension the tape, i.e. forward space to end of tape and rewind.
+ */
+int
+tape_std_mtreten(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+	int rc;
+
+	request = tape_alloc_request(4, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_FSF;
+	/* setup ccws */
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_cc(request->cpaddr + 1,FORSPACEFILE, 0, NULL);
+	tape_ccw_cc(request->cpaddr + 2, NOP, 0, NULL);
+	tape_ccw_end(request->cpaddr + 3, CCW_CMD_TIC, 0, request->cpaddr);
+	/* execute it, MTRETEN rc gets ignored */
+	rc = tape_do_io_interruptible(device, request);
+	tape_free_request(request);
+	return tape_mtop(device, MTREW, 1);
+}
+
+/*
+ * MTERASE: erases the tape.
+ */
+int
+tape_std_mterase(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+
+	request = tape_alloc_request(6, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_DSE;
+	/* setup ccws */
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_cc(request->cpaddr + 1, REWIND, 0, NULL);
+	tape_ccw_cc(request->cpaddr + 2, ERASE_GAP, 0, NULL);
+	tape_ccw_cc(request->cpaddr + 3, DATA_SEC_ERASE, 0, NULL);
+	tape_ccw_cc(request->cpaddr + 4, REWIND, 0, NULL);
+	tape_ccw_end(request->cpaddr + 5, NOP, 0, NULL);
+
+	/* execute it */
+	return tape_do_io_free(device, request);
+}
+
+/*
+ * MTUNLOAD: Rewind the tape and unload it.
+ */
+int
+tape_std_mtunload(struct tape_device *device, int mt_count)
+{
+	return tape_mtop(device, MTOFFL, mt_count);
+}
+
+/*
+ * MTCOMPRESSION: used to enable compression.
+ * Sets the IDRC on/off.
+ */
+int
+tape_std_mtcompression(struct tape_device *device, int mt_count)
+{
+	struct tape_request *request;
+
+	if (mt_count < 0 || mt_count > 1) {
+		DBF_EXCEPTION(6, "xcom parm\n");
+		return -EINVAL;
+	}
+	request = tape_alloc_request(2, 0);
+	if (IS_ERR(request))
+		return PTR_ERR(request);
+	request->op = TO_NOP;
+	/* setup ccws */
+	if (mt_count == 0)
+		*device->modeset_byte &= ~0x08;
+	else
+		*device->modeset_byte |= 0x08;
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_end(request->cpaddr + 1, NOP, 0, NULL);
+	/* execute it */
+	return tape_do_io_free(device, request);
+}
+
+/*
+ * Read Block
+ */
+struct tape_request *
+tape_std_read_block(struct tape_device *device, size_t count)
+{
+	struct tape_request *request;
+
+	/*
+	 * We have to alloc 4 ccws in order to be able to transform request
+	 * into a read backward request in error case.
+	 */
+	request = tape_alloc_request(4, 0);
+	if (IS_ERR(request)) {
+		DBF_EXCEPTION(6, "xrbl fail");
+		return request;
+	}
+	request->op = TO_RFO;
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_end_idal(request->cpaddr + 1, READ_FORWARD,
+			  device->char_data.idal_buf);
+	DBF_EVENT(6, "xrbl ccwg\n");
+	return request;
+}
+
+/*
+ * Read Block backward transformation function.
+ */
+void
+tape_std_read_backward(struct tape_device *device, struct tape_request *request)
+{
+	/*
+	 * We have allocated 4 ccws in tape_std_read, so we can now
+	 * transform the request to a read backward, followed by a
+	 * forward space block.
+	 */
+	request->op = TO_RBA;
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_cc_idal(request->cpaddr + 1, READ_BACKWARD,
+			 device->char_data.idal_buf);
+	tape_ccw_cc(request->cpaddr + 2, FORSPACEBLOCK, 0, NULL);
+	tape_ccw_end(request->cpaddr + 3, NOP, 0, NULL);
+	DBF_EVENT(6, "xrop ccwg");}
+
+/*
+ * Write Block
+ */
+struct tape_request *
+tape_std_write_block(struct tape_device *device, size_t count)
+{
+	struct tape_request *request;
+
+	request = tape_alloc_request(2, 0);
+	if (IS_ERR(request)) {
+		DBF_EXCEPTION(6, "xwbl fail\n");
+		return request;
+	}
+	request->op = TO_WRI;
+	tape_ccw_cc(request->cpaddr, MODE_SET_DB, 1, device->modeset_byte);
+	tape_ccw_end_idal(request->cpaddr + 1, WRITE_CMD,
+			  device->char_data.idal_buf);
+	DBF_EVENT(6, "xwbl ccwg\n");
+	return request;
+}
+
+/*
+ * This routine is called by frontend after an ENOSP on write
+ */
+void
+tape_std_process_eov(struct tape_device *device)
+{
+	/*
+	 * End of volume: We have to backspace the last written record, then
+	 * we TRY to write a tapemark and then backspace over the written TM
+	 */
+	if (tape_mtop(device, MTBSR, 1) == 0 &&
+	    tape_mtop(device, MTWEOF, 1) == 0) {
+		tape_mtop(device, MTBSR, 1);
+	}
+}
+
+EXPORT_SYMBOL(tape_std_assign);
+EXPORT_SYMBOL(tape_std_unassign);
+EXPORT_SYMBOL(tape_std_display);
+EXPORT_SYMBOL(tape_std_read_block_id);
+EXPORT_SYMBOL(tape_std_mtload);
+EXPORT_SYMBOL(tape_std_mtsetblk);
+EXPORT_SYMBOL(tape_std_mtreset);
+EXPORT_SYMBOL(tape_std_mtfsf);
+EXPORT_SYMBOL(tape_std_mtfsr);
+EXPORT_SYMBOL(tape_std_mtbsr);
+EXPORT_SYMBOL(tape_std_mtweof);
+EXPORT_SYMBOL(tape_std_mtbsfm);
+EXPORT_SYMBOL(tape_std_mtbsf);
+EXPORT_SYMBOL(tape_std_mtfsfm);
+EXPORT_SYMBOL(tape_std_mtrew);
+EXPORT_SYMBOL(tape_std_mtoffl);
+EXPORT_SYMBOL(tape_std_mtnop);
+EXPORT_SYMBOL(tape_std_mteom);
+EXPORT_SYMBOL(tape_std_mtreten);
+EXPORT_SYMBOL(tape_std_mterase);
+EXPORT_SYMBOL(tape_std_mtunload);
+EXPORT_SYMBOL(tape_std_mtcompression);
+EXPORT_SYMBOL(tape_std_read_block);
+EXPORT_SYMBOL(tape_std_read_backward);
+EXPORT_SYMBOL(tape_std_write_block);
+EXPORT_SYMBOL(tape_std_process_eov);

kernel/drivers/s390/char/tape_std.h

@@ -0,0 +1,159 @@
+/*
+ *  drivers/s390/char/tape_std.h
+ *    standard tape device functions for ibm tapes.
+ *
+ *    Copyright (C) IBM Corp. 2001,2006
+ *    Author(s): Carsten Otte <cotte@de.ibm.com>
+ *		 Tuan Ngo-Anh <ngoanh@de.ibm.com>
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>
+ */
+
+#ifndef _TAPE_STD_H
+#define _TAPE_STD_H
+
+#include <asm/tape390.h>
+
+/*
+ * Biggest block size to handle. Currently 64K because we only build
+ * channel programs without data chaining.
+ */
+#define MAX_BLOCKSIZE   65535
+
+/*
+ * The CCW commands for the Tape type of command.
+ */
+#define INVALID_00		0x00	/* Invalid cmd */
+#define BACKSPACEBLOCK		0x27	/* Back Space block */
+#define BACKSPACEFILE		0x2f	/* Back Space file */
+#define DATA_SEC_ERASE		0x97	/* Data security erase */
+#define ERASE_GAP		0x17	/* Erase Gap */
+#define FORSPACEBLOCK		0x37	/* Forward space block */
+#define FORSPACEFILE		0x3F	/* Forward Space file */
+#define FORCE_STREAM_CNT	0xEB	/* Forced streaming count # */
+#define NOP			0x03	/* No operation	*/
+#define READ_FORWARD		0x02	/* Read forward */
+#define REWIND			0x07	/* Rewind */
+#define REWIND_UNLOAD		0x0F	/* Rewind and Unload */
+#define SENSE			0x04	/* Sense */
+#define NEW_MODE_SET		0xEB	/* Guess it is Mode set */
+#define WRITE_CMD		0x01	/* Write */
+#define WRITETAPEMARK		0x1F	/* Write Tape Mark */
+
+#define ASSIGN			0xB7	/* 3420 REJECT,3480 OK	*/
+#define CONTROL_ACCESS		0xE3	/* Set high speed */
+#define DIAG_MODE_SET		0x0B	/* 3420 NOP, 3480 REJECT */
+#define LOAD_DISPLAY		0x9F	/* 3420 REJECT,3480 OK */
+#define LOCATE			0x4F	/* 3420 REJ, 3480 NOP */
+#define LOOP_WRITE_TO_READ	0x8B	/* 3480 REJECT */
+#define MODE_SET_DB		0xDB	/* 3420 REJECT,3480 OK */
+#define MODE_SET_C3		0xC3	/* for 3420 */
+#define MODE_SET_CB		0xCB	/* for 3420 */
+#define MODE_SET_D3		0xD3	/* for 3420 */
+#define READ_BACKWARD		0x0C	/* */
+#define READ_BLOCK_ID		0x22	/* 3420 REJECT,3480 OK */
+#define READ_BUFFER		0x12	/* 3420 REJECT,3480 OK */
+#define READ_BUFF_LOG		0x24	/* 3420 REJECT,3480 OK */
+#define RELEASE			0xD4	/* 3420 NOP, 3480 REJECT */
+#define REQ_TRK_IN_ERROR	0x1B	/* 3420 NOP, 3480 REJECT */
+#define RESERVE			0xF4	/* 3420 NOP, 3480 REJECT */
+#define SENSE_GROUP_ID		0x34	/* 3420 REJECT,3480 OK */
+#define SENSE_ID		0xE4	/* 3420 REJECT,3480 OK */
+#define READ_DEV_CHAR		0x64	/* Read device characteristics */
+#define SET_DIAGNOSE		0x4B	/* 3420 NOP, 3480 REJECT */
+#define SET_GROUP_ID		0xAF	/* 3420 REJECT,3480 OK */
+#define SET_TAPE_WRITE_IMMED	0xC3	/* for 3480 */
+#define SUSPEND			0x5B	/* 3420 REJ, 3480 NOP */
+#define SYNC			0x43	/* Synchronize (flush buffer) */
+#define UNASSIGN		0xC7	/* 3420 REJECT,3480 OK */
+#define PERF_SUBSYS_FUNC	0x77	/* 3490 CMD */
+#define READ_CONFIG_DATA	0xFA	/* 3490 CMD */
+#define READ_MESSAGE_ID		0x4E	/* 3490 CMD */
+#define READ_SUBSYS_DATA	0x3E	/* 3490 CMD */
+#define SET_INTERFACE_ID	0x73	/* 3490 CMD */
+
+#define SENSE_COMMAND_REJECT		0x80
+#define SENSE_INTERVENTION_REQUIRED	0x40
+#define SENSE_BUS_OUT_CHECK		0x20
+#define SENSE_EQUIPMENT_CHECK		0x10
+#define SENSE_DATA_CHECK		0x08
+#define SENSE_OVERRUN			0x04
+#define SENSE_DEFERRED_UNIT_CHECK	0x02
+#define SENSE_ASSIGNED_ELSEWHERE	0x01
+
+#define SENSE_LOCATE_FAILURE		0x80
+#define SENSE_DRIVE_ONLINE		0x40
+#define SENSE_RESERVED			0x20
+#define SENSE_RECORD_SEQUENCE_ERR	0x10
+#define SENSE_BEGINNING_OF_TAPE		0x08
+#define SENSE_WRITE_MODE		0x04
+#define SENSE_WRITE_PROTECT		0x02
+#define SENSE_NOT_CAPABLE		0x01
+
+#define SENSE_CHANNEL_ADAPTER_CODE	0xE0
+#define SENSE_CHANNEL_ADAPTER_LOC	0x10
+#define SENSE_REPORTING_CU		0x08
+#define SENSE_AUTOMATIC_LOADER		0x04
+#define SENSE_TAPE_SYNC_MODE		0x02
+#define SENSE_TAPE_POSITIONING		0x01
+
+/* discipline functions */
+struct tape_request *tape_std_read_block(struct tape_device *, size_t);
+void tape_std_read_backward(struct tape_device *device,
+			    struct tape_request *request);
+struct tape_request *tape_std_write_block(struct tape_device *, size_t);
+struct tape_request *tape_std_bread(struct tape_device *, struct request *);
+void tape_std_free_bread(struct tape_request *);
+void tape_std_check_locate(struct tape_device *, struct tape_request *);
+struct tape_request *tape_std_bwrite(struct request *,
+				     struct tape_device *, int);
+
+/* Some non-mtop commands. */
+int tape_std_assign(struct tape_device *);
+int tape_std_unassign(struct tape_device *);
+int tape_std_read_block_id(struct tape_device *device, __u64 *id);
+int tape_std_display(struct tape_device *, struct display_struct *disp);
+int tape_std_terminate_write(struct tape_device *);
+
+/* Standard magnetic tape commands. */
+int tape_std_mtbsf(struct tape_device *, int);
+int tape_std_mtbsfm(struct tape_device *, int);
+int tape_std_mtbsr(struct tape_device *, int);
+int tape_std_mtcompression(struct tape_device *, int);
+int tape_std_mteom(struct tape_device *, int);
+int tape_std_mterase(struct tape_device *, int);
+int tape_std_mtfsf(struct tape_device *, int);
+int tape_std_mtfsfm(struct tape_device *, int);
+int tape_std_mtfsr(struct tape_device *, int);
+int tape_std_mtload(struct tape_device *, int);
+int tape_std_mtnop(struct tape_device *, int);
+int tape_std_mtoffl(struct tape_device *, int);
+int tape_std_mtreset(struct tape_device *, int);
+int tape_std_mtreten(struct tape_device *, int);
+int tape_std_mtrew(struct tape_device *, int);
+int tape_std_mtsetblk(struct tape_device *, int);
+int tape_std_mtunload(struct tape_device *, int);
+int tape_std_mtweof(struct tape_device *, int);
+
+/* Event handlers */
+void tape_std_default_handler(struct tape_device *);
+void tape_std_unexpect_uchk_handler(struct tape_device *);
+void tape_std_irq(struct tape_device *);
+void tape_std_process_eov(struct tape_device *);
+
+// the error recovery stuff:
+void tape_std_error_recovery(struct tape_device *);
+void tape_std_error_recovery_has_failed(struct tape_device *,int error_id);
+void tape_std_error_recovery_succeded(struct tape_device *);
+void tape_std_error_recovery_do_retry(struct tape_device *);
+void tape_std_error_recovery_read_opposite(struct tape_device *);
+void tape_std_error_recovery_HWBUG(struct tape_device *, int condno);
+
+/* S390 tape types */
+enum s390_tape_type {
+        tape_3480,
+        tape_3490,
+        tape_3590,
+        tape_3592,
+};
+
+#endif // _TAPE_STD_H

kernel/drivers/s390/char/tty3270.h

@@ -0,0 +1,16 @@
+/*
+ *  drivers/s390/char/tty3270.h
+ *
+ *    Copyright IBM Corp. 2007
+ *
+ */
+
+#ifndef __DRIVERS_S390_CHAR_TTY3270_H
+#define __DRIVERS_S390_CHAR_TTY3270_H
+
+#include <linux/tty.h>
+#include <linux/tty_driver.h>
+
+extern struct tty_driver *tty3270_driver;
+
+#endif /* __DRIVERS_S390_CHAR_TTY3270_H */

kernel/drivers/s390/char/vmcp.c

@@ -0,0 +1,225 @@
+/*
+ * Copyright IBM Corp. 2004,2007
+ * Interface implementation for communication with the z/VM control program
+ * Author(s): Christian Borntraeger <borntraeger@de.ibm.com>
+ *
+ *
+ * z/VMs CP offers the possibility to issue commands via the diagnose code 8
+ * this driver implements a character device that issues these commands and
+ * returns the answer of CP.
+
+ * The idea of this driver is based on cpint from Neale Ferguson and #CP in CMS
+ */
+
+#define KMSG_COMPONENT "vmcp"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <asm/cpcmd.h>
+#include <asm/debug.h>
+#include <asm/uaccess.h>
+#include "vmcp.h"
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Christian Borntraeger <borntraeger@de.ibm.com>");
+MODULE_DESCRIPTION("z/VM CP interface");
+
+static debug_info_t *vmcp_debug;
+
+static int vmcp_open(struct inode *inode, struct file *file)
+{
+	struct vmcp_session *session;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	session = kmalloc(sizeof(*session), GFP_KERNEL);
+	if (!session)
+		return -ENOMEM;
+
+	session->bufsize = PAGE_SIZE;
+	session->response = NULL;
+	session->resp_size = 0;
+	mutex_init(&session->mutex);
+	file->private_data = session;
+	return nonseekable_open(inode, file);
+}
+
+static int vmcp_release(struct inode *inode, struct file *file)
+{
+	struct vmcp_session *session;
+
+	session = (struct vmcp_session *)file->private_data;
+	file->private_data = NULL;
+	free_pages((unsigned long)session->response, get_order(session->bufsize));
+	kfree(session);
+	return 0;
+}
+
+static ssize_t
+vmcp_read(struct file *file, char __user *buff, size_t count, loff_t *ppos)
+{
+	ssize_t ret;
+	size_t size;
+	struct vmcp_session *session;
+
+	session = file->private_data;
+	if (mutex_lock_interruptible(&session->mutex))
+		return -ERESTARTSYS;
+	if (!session->response) {
+		mutex_unlock(&session->mutex);
+		return 0;
+	}
+	size = min_t(size_t, session->resp_size, session->bufsize);
+	ret = simple_read_from_buffer(buff, count, ppos,
+					session->response, size);
+
+	mutex_unlock(&session->mutex);
+
+	return ret;
+}
+
+static ssize_t
+vmcp_write(struct file *file, const char __user *buff, size_t count,
+	   loff_t *ppos)
+{
+	char *cmd;
+	struct vmcp_session *session;
+
+	if (count > 240)
+		return -EINVAL;
+	cmd = kmalloc(count + 1, GFP_KERNEL);
+	if (!cmd)
+		return -ENOMEM;
+	if (copy_from_user(cmd, buff, count)) {
+		kfree(cmd);
+		return -EFAULT;
+	}
+	cmd[count] = '\0';
+	session = (struct vmcp_session *)file->private_data;
+	if (mutex_lock_interruptible(&session->mutex)) {
+		kfree(cmd);
+		return -ERESTARTSYS;
+	}
+	if (!session->response)
+		session->response = (char *)__get_free_pages(GFP_KERNEL
+						| __GFP_REPEAT | GFP_DMA,
+						get_order(session->bufsize));
+	if (!session->response) {
+		mutex_unlock(&session->mutex);
+		kfree(cmd);
+		return -ENOMEM;
+	}
+	debug_text_event(vmcp_debug, 1, cmd);
+	session->resp_size = cpcmd(cmd, session->response, session->bufsize,
+				   &session->resp_code);
+	mutex_unlock(&session->mutex);
+	kfree(cmd);
+	*ppos = 0;		/* reset the file pointer after a command */
+	return count;
+}
+
+
+/*
+ * These ioctls are available, as the semantics of the diagnose 8 call
+ * does not fit very well into a Linux call. Diagnose X'08' is described in
+ * CP Programming Services SC24-6084-00
+ *
+ * VMCP_GETCODE: gives the CP return code back to user space
+ * VMCP_SETBUF: sets the response buffer for the next write call. diagnose 8
+ * expects adjacent pages in real storage and to make matters worse, we
+ * dont know the size of the response. Therefore we default to PAGESIZE and
+ * let userspace to change the response size, if userspace expects a bigger
+ * response
+ */
+static long vmcp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct vmcp_session *session;
+	int temp;
+
+	session = (struct vmcp_session *)file->private_data;
+	if (mutex_lock_interruptible(&session->mutex))
+		return -ERESTARTSYS;
+	switch (cmd) {
+	case VMCP_GETCODE:
+		temp = session->resp_code;
+		mutex_unlock(&session->mutex);
+		return put_user(temp, (int __user *)arg);
+	case VMCP_SETBUF:
+		free_pages((unsigned long)session->response,
+				get_order(session->bufsize));
+		session->response=NULL;
+		temp = get_user(session->bufsize, (int __user *)arg);
+		if (get_order(session->bufsize) > 8) {
+			session->bufsize = PAGE_SIZE;
+			temp = -EINVAL;
+		}
+		mutex_unlock(&session->mutex);
+		return temp;
+	case VMCP_GETSIZE:
+		temp = session->resp_size;
+		mutex_unlock(&session->mutex);
+		return put_user(temp, (int __user *)arg);
+	default:
+		mutex_unlock(&session->mutex);
+		return -ENOIOCTLCMD;
+	}
+}
+
+static const struct file_operations vmcp_fops = {
+	.owner		= THIS_MODULE,
+	.open		= vmcp_open,
+	.release	= vmcp_release,
+	.read		= vmcp_read,
+	.write		= vmcp_write,
+	.unlocked_ioctl	= vmcp_ioctl,
+	.compat_ioctl	= vmcp_ioctl,
+};
+
+static struct miscdevice vmcp_dev = {
+	.name	= "vmcp",
+	.minor	= MISC_DYNAMIC_MINOR,
+	.fops	= &vmcp_fops,
+};
+
+static int __init vmcp_init(void)
+{
+	int ret;
+
+	if (!MACHINE_IS_VM) {
+		pr_warning("The z/VM CP interface device driver cannot be "
+			   "loaded without z/VM\n");
+		return -ENODEV;
+	}
+
+	vmcp_debug = debug_register("vmcp", 1, 1, 240);
+	if (!vmcp_debug)
+		return -ENOMEM;
+
+	ret = debug_register_view(vmcp_debug, &debug_hex_ascii_view);
+	if (ret) {
+		debug_unregister(vmcp_debug);
+		return ret;
+	}
+
+	ret = misc_register(&vmcp_dev);
+	if (ret) {
+		debug_unregister(vmcp_debug);
+		return ret;
+	}
+
+	return 0;
+}
+
+static void __exit vmcp_exit(void)
+{
+	misc_deregister(&vmcp_dev);
+	debug_unregister(vmcp_debug);
+}
+
+module_init(vmcp_init);
+module_exit(vmcp_exit);

kernel/drivers/s390/char/vmcp.h

@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2004, 2005 IBM Corporation
+ * Interface implementation for communication with the z/VM control program
+ * Version 1.0
+ * Author(s): Christian Borntraeger <cborntra@de.ibm.com>
+ *
+ *
+ * z/VMs CP offers the possibility to issue commands via the diagnose code 8
+ * this driver implements a character device that issues these commands and
+ * returns the answer of CP.
+ *
+ * The idea of this driver is based on cpint from Neale Ferguson
+ */
+
+#include <linux/ioctl.h>
+#include <linux/mutex.h>
+
+#define VMCP_GETCODE _IOR(0x10, 1, int)
+#define VMCP_SETBUF _IOW(0x10, 2, int)
+#define VMCP_GETSIZE _IOR(0x10, 3, int)
+
+struct vmcp_session {
+	unsigned int bufsize;
+	char *response;
+	int resp_size;
+	int resp_code;
+	/* As we use copy_from/to_user, which might     *
+	 * sleep and cannot use a spinlock              */
+	struct mutex mutex;
+};

kernel/drivers/s390/char/vmlogrdr.c

@@ -0,0 +1,902 @@
+/*
+ * drivers/s390/char/vmlogrdr.c
+ *	character device driver for reading z/VM system service records
+ *
+ *
+ *	Copyright IBM Corp. 2004, 2009
+ *	character device driver for reading z/VM system service records,
+ *	Version 1.0
+ *	Author(s): Xenia Tkatschow <xenia@us.ibm.com>
+ *		   Stefan Weinhuber <wein@de.ibm.com>
+ *
+ */
+
+#define KMSG_COMPONENT "vmlogrdr"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <asm/atomic.h>
+#include <asm/uaccess.h>
+#include <asm/cpcmd.h>
+#include <asm/debug.h>
+#include <asm/ebcdic.h>
+#include <net/iucv/iucv.h>
+#include <linux/kmod.h>
+#include <linux/cdev.h>
+#include <linux/device.h>
+#include <linux/smp_lock.h>
+#include <linux/string.h>
+
+MODULE_AUTHOR
+	("(C) 2004 IBM Corporation by Xenia Tkatschow (xenia@us.ibm.com)\n"
+	 "                            Stefan Weinhuber (wein@de.ibm.com)");
+MODULE_DESCRIPTION ("Character device driver for reading z/VM "
+		    "system service records.");
+MODULE_LICENSE("GPL");
+
+
+/*
+ * The size of the buffer for iucv data transfer is one page,
+ * but in addition to the data we read from iucv we also
+ * place an integer and some characters into that buffer,
+ * so the maximum size for record data is a little less then
+ * one page.
+ */
+#define NET_BUFFER_SIZE	(PAGE_SIZE - sizeof(int) - sizeof(FENCE))
+
+/*
+ * The elements that are concurrently accessed by bottom halves are
+ * connection_established, iucv_path_severed, local_interrupt_buffer
+ * and receive_ready. The first three can be protected by
+ * priv_lock.  receive_ready is atomic, so it can be incremented and
+ * decremented without holding a lock.
+ * The variable dev_in_use needs to be protected by the lock, since
+ * it's a flag used by open to make sure that the device is opened only
+ * by one user at the same time.
+ */
+struct vmlogrdr_priv_t {
+	char system_service[8];
+	char internal_name[8];
+	char recording_name[8];
+	struct iucv_path *path;
+	int connection_established;
+	int iucv_path_severed;
+	struct iucv_message local_interrupt_buffer;
+	atomic_t receive_ready;
+	int minor_num;
+	char * buffer;
+	char * current_position;
+	int remaining;
+	ulong residual_length;
+	int buffer_free;
+	int dev_in_use; /* 1: already opened, 0: not opened*/
+	spinlock_t priv_lock;
+	struct device  *device;
+	struct device  *class_device;
+	int autorecording;
+	int autopurge;
+};
+
+
+/*
+ * File operation structure for vmlogrdr devices
+ */
+static int vmlogrdr_open(struct inode *, struct file *);
+static int vmlogrdr_release(struct inode *, struct file *);
+static ssize_t vmlogrdr_read (struct file *filp, char __user *data,
+			      size_t count, loff_t * ppos);
+
+static const struct file_operations vmlogrdr_fops = {
+	.owner   = THIS_MODULE,
+	.open    = vmlogrdr_open,
+	.release = vmlogrdr_release,
+	.read    = vmlogrdr_read,
+};
+
+
+static void vmlogrdr_iucv_path_complete(struct iucv_path *, u8 ipuser[16]);
+static void vmlogrdr_iucv_path_severed(struct iucv_path *, u8 ipuser[16]);
+static void vmlogrdr_iucv_message_pending(struct iucv_path *,
+					  struct iucv_message *);
+
+
+static struct iucv_handler vmlogrdr_iucv_handler = {
+	.path_complete	 = vmlogrdr_iucv_path_complete,
+	.path_severed	 = vmlogrdr_iucv_path_severed,
+	.message_pending = vmlogrdr_iucv_message_pending,
+};
+
+
+static DECLARE_WAIT_QUEUE_HEAD(conn_wait_queue);
+static DECLARE_WAIT_QUEUE_HEAD(read_wait_queue);
+
+/*
+ * pointer to system service private structure
+ * minor number 0 --> logrec
+ * minor number 1 --> account
+ * minor number 2 --> symptom
+ */
+
+static struct vmlogrdr_priv_t sys_ser[] = {
+	{ .system_service = "*LOGREC ",
+	  .internal_name  = "logrec",
+	  .recording_name = "EREP",
+	  .minor_num      = 0,
+	  .buffer_free    = 1,
+	  .priv_lock	  = __SPIN_LOCK_UNLOCKED(sys_ser[0].priv_lock),
+	  .autorecording  = 1,
+	  .autopurge      = 1,
+	},
+	{ .system_service = "*ACCOUNT",
+	  .internal_name  = "account",
+	  .recording_name = "ACCOUNT",
+	  .minor_num      = 1,
+	  .buffer_free    = 1,
+	  .priv_lock	  = __SPIN_LOCK_UNLOCKED(sys_ser[1].priv_lock),
+	  .autorecording  = 1,
+	  .autopurge      = 1,
+	},
+	{ .system_service = "*SYMPTOM",
+	  .internal_name  = "symptom",
+	  .recording_name = "SYMPTOM",
+	  .minor_num      = 2,
+	  .buffer_free    = 1,
+	  .priv_lock	  = __SPIN_LOCK_UNLOCKED(sys_ser[2].priv_lock),
+	  .autorecording  = 1,
+	  .autopurge      = 1,
+	}
+};
+
+#define MAXMINOR  (sizeof(sys_ser)/sizeof(struct vmlogrdr_priv_t))
+
+static char FENCE[] = {"EOR"};
+static int vmlogrdr_major = 0;
+static struct cdev  *vmlogrdr_cdev = NULL;
+static int recording_class_AB;
+
+
+static void vmlogrdr_iucv_path_complete(struct iucv_path *path, u8 ipuser[16])
+{
+	struct vmlogrdr_priv_t * logptr = path->private;
+
+	spin_lock(&logptr->priv_lock);
+	logptr->connection_established = 1;
+	spin_unlock(&logptr->priv_lock);
+	wake_up(&conn_wait_queue);
+}
+
+
+static void vmlogrdr_iucv_path_severed(struct iucv_path *path, u8 ipuser[16])
+{
+	struct vmlogrdr_priv_t * logptr = path->private;
+	u8 reason = (u8) ipuser[8];
+
+	pr_err("vmlogrdr: connection severed with reason %i\n", reason);
+
+	iucv_path_sever(path, NULL);
+	kfree(path);
+	logptr->path = NULL;
+
+	spin_lock(&logptr->priv_lock);
+	logptr->connection_established = 0;
+	logptr->iucv_path_severed = 1;
+	spin_unlock(&logptr->priv_lock);
+
+	wake_up(&conn_wait_queue);
+	/* just in case we're sleeping waiting for a record */
+	wake_up_interruptible(&read_wait_queue);
+}
+
+
+static void vmlogrdr_iucv_message_pending(struct iucv_path *path,
+					  struct iucv_message *msg)
+{
+	struct vmlogrdr_priv_t * logptr = path->private;
+
+	/*
+	 * This function is the bottom half so it should be quick.
+	 * Copy the external interrupt data into our local eib and increment
+	 * the usage count
+	 */
+	spin_lock(&logptr->priv_lock);
+	memcpy(&logptr->local_interrupt_buffer, msg, sizeof(*msg));
+	atomic_inc(&logptr->receive_ready);
+	spin_unlock(&logptr->priv_lock);
+	wake_up_interruptible(&read_wait_queue);
+}
+
+
+static int vmlogrdr_get_recording_class_AB(void)
+{
+	char cp_command[]="QUERY COMMAND RECORDING ";
+	char cp_response[80];
+	char *tail;
+	int len,i;
+
+	cpcmd(cp_command, cp_response, sizeof(cp_response), NULL);
+	len = strnlen(cp_response,sizeof(cp_response));
+	// now the parsing
+	tail=strnchr(cp_response,len,'=');
+	if (!tail)
+		return 0;
+	tail++;
+	if (!strncmp("ANY",tail,3))
+		return 1;
+	if (!strncmp("NONE",tail,4))
+		return 0;
+	/*
+	 * expect comma separated list of classes here, if one of them
+	 * is A or B return 1 otherwise 0
+	 */
+        for (i=tail-cp_response; i<len; i++)
+		if ( cp_response[i]=='A' || cp_response[i]=='B' )
+			return 1;
+	return 0;
+}
+
+
+static int vmlogrdr_recording(struct vmlogrdr_priv_t * logptr,
+			      int action, int purge)
+{
+
+	char cp_command[80];
+	char cp_response[160];
+	char *onoff, *qid_string;
+
+	memset(cp_command, 0x00, sizeof(cp_command));
+	memset(cp_response, 0x00, sizeof(cp_response));
+
+        onoff = ((action == 1) ? "ON" : "OFF");
+	qid_string = ((recording_class_AB == 1) ? " QID * " : "");
+
+        /*
+	 * The recording commands needs to be called with option QID
+	 * for guests that have previlege classes A or B.
+	 * Purging has to be done as separate step, because recording
+	 * can't be switched on as long as records are on the queue.
+	 * Doing both at the same time doesn't work.
+	 */
+
+	if (purge) {
+		snprintf(cp_command, sizeof(cp_command),
+			 "RECORDING %s PURGE %s",
+			 logptr->recording_name,
+			 qid_string);
+
+		cpcmd(cp_command, cp_response, sizeof(cp_response), NULL);
+	}
+
+	memset(cp_command, 0x00, sizeof(cp_command));
+	memset(cp_response, 0x00, sizeof(cp_response));
+	snprintf(cp_command, sizeof(cp_command), "RECORDING %s %s %s",
+		logptr->recording_name,
+		onoff,
+		qid_string);
+
+	cpcmd(cp_command, cp_response, sizeof(cp_response), NULL);
+	/* The recording command will usually answer with 'Command complete'
+	 * on success, but when the specific service was never connected
+	 * before then there might be an additional informational message
+	 * 'HCPCRC8072I Recording entry not found' before the
+         * 'Command complete'. So I use strstr rather then the strncmp.
+	 */
+	if (strstr(cp_response,"Command complete"))
+		return 0;
+	else
+		return -EIO;
+
+}
+
+
+static int vmlogrdr_open (struct inode *inode, struct file *filp)
+{
+	int dev_num = 0;
+	struct vmlogrdr_priv_t * logptr = NULL;
+	int connect_rc = 0;
+	int ret;
+
+	dev_num = iminor(inode);
+	if (dev_num > MAXMINOR)
+		return -ENODEV;
+	logptr = &sys_ser[dev_num];
+
+	/*
+	 * only allow for blocking reads to be open
+	 */
+	if (filp->f_flags & O_NONBLOCK)
+		return -ENOSYS;
+
+	/* Besure this device hasn't already been opened */
+	lock_kernel();
+	spin_lock_bh(&logptr->priv_lock);
+	if (logptr->dev_in_use)	{
+		spin_unlock_bh(&logptr->priv_lock);
+		unlock_kernel();
+		return -EBUSY;
+	}
+	logptr->dev_in_use = 1;
+	logptr->connection_established = 0;
+	logptr->iucv_path_severed = 0;
+	atomic_set(&logptr->receive_ready, 0);
+	logptr->buffer_free = 1;
+	spin_unlock_bh(&logptr->priv_lock);
+
+	/* set the file options */
+	filp->private_data = logptr;
+	filp->f_op = &vmlogrdr_fops;
+
+	/* start recording for this service*/
+	if (logptr->autorecording) {
+		ret = vmlogrdr_recording(logptr,1,logptr->autopurge);
+		if (ret)
+			pr_warning("vmlogrdr: failed to start "
+				   "recording automatically\n");
+	}
+
+	/* create connection to the system service */
+	logptr->path = iucv_path_alloc(10, 0, GFP_KERNEL);
+	if (!logptr->path)
+		goto out_dev;
+	connect_rc = iucv_path_connect(logptr->path, &vmlogrdr_iucv_handler,
+				       logptr->system_service, NULL, NULL,
+				       logptr);
+	if (connect_rc) {
+		pr_err("vmlogrdr: iucv connection to %s "
+		       "failed with rc %i \n",
+		       logptr->system_service, connect_rc);
+		goto out_path;
+	}
+
+	/* We've issued the connect and now we must wait for a
+	 * ConnectionComplete or ConnectinSevered Interrupt
+	 * before we can continue to process.
+	 */
+	wait_event(conn_wait_queue, (logptr->connection_established)
+		   || (logptr->iucv_path_severed));
+	if (logptr->iucv_path_severed)
+		goto out_record;
+ 	ret = nonseekable_open(inode, filp);
+	unlock_kernel();
+	return ret;
+
+out_record:
+	if (logptr->autorecording)
+		vmlogrdr_recording(logptr,0,logptr->autopurge);
+out_path:
+	kfree(logptr->path);	/* kfree(NULL) is ok. */
+	logptr->path = NULL;
+out_dev:
+	logptr->dev_in_use = 0;
+	unlock_kernel();
+	return -EIO;
+}
+
+
+static int vmlogrdr_release (struct inode *inode, struct file *filp)
+{
+	int ret;
+
+	struct vmlogrdr_priv_t * logptr = filp->private_data;
+
+	iucv_path_sever(logptr->path, NULL);
+	kfree(logptr->path);
+	logptr->path = NULL;
+	if (logptr->autorecording) {
+		ret = vmlogrdr_recording(logptr,0,logptr->autopurge);
+		if (ret)
+			pr_warning("vmlogrdr: failed to stop "
+				   "recording automatically\n");
+	}
+	logptr->dev_in_use = 0;
+
+	return 0;
+}
+
+
+static int vmlogrdr_receive_data(struct vmlogrdr_priv_t *priv)
+{
+	int rc, *temp;
+	/* we need to keep track of two data sizes here:
+	 * The number of bytes we need to receive from iucv and
+	 * the total number of bytes we actually write into the buffer.
+	 */
+	int user_data_count, iucv_data_count;
+	char * buffer;
+
+	if (atomic_read(&priv->receive_ready)) {
+		spin_lock_bh(&priv->priv_lock);
+		if (priv->residual_length){
+			/* receive second half of a record */
+			iucv_data_count = priv->residual_length;
+			user_data_count = 0;
+			buffer = priv->buffer;
+		} else {
+			/* receive a new record:
+			 * We need to return the total length of the record
+                         * + size of FENCE in the first 4 bytes of the buffer.
+		         */
+			iucv_data_count = priv->local_interrupt_buffer.length;
+			user_data_count = sizeof(int);
+			temp = (int*)priv->buffer;
+			*temp= iucv_data_count + sizeof(FENCE);
+			buffer = priv->buffer + sizeof(int);
+		}
+		/*
+		 * If the record is bigger than our buffer, we receive only
+		 * a part of it. We can get the rest later.
+		 */
+		if (iucv_data_count > NET_BUFFER_SIZE)
+			iucv_data_count = NET_BUFFER_SIZE;
+		rc = iucv_message_receive(priv->path,
+					  &priv->local_interrupt_buffer,
+					  0, buffer, iucv_data_count,
+					  &priv->residual_length);
+		spin_unlock_bh(&priv->priv_lock);
+		/* An rc of 5 indicates that the record was bigger than
+		 * the buffer, which is OK for us. A 9 indicates that the
+		 * record was purged befor we could receive it.
+		 */
+		if (rc == 5)
+			rc = 0;
+		if (rc == 9)
+			atomic_set(&priv->receive_ready, 0);
+	} else {
+		rc = 1;
+	}
+	if (!rc) {
+		priv->buffer_free = 0;
+ 		user_data_count += iucv_data_count;
+		priv->current_position = priv->buffer;
+		if (priv->residual_length == 0){
+			/* the whole record has been captured,
+			 * now add the fence */
+			atomic_dec(&priv->receive_ready);
+			buffer = priv->buffer + user_data_count;
+			memcpy(buffer, FENCE, sizeof(FENCE));
+			user_data_count += sizeof(FENCE);
+		}
+		priv->remaining = user_data_count;
+	}
+
+	return rc;
+}
+
+
+static ssize_t vmlogrdr_read(struct file *filp, char __user *data,
+			     size_t count, loff_t * ppos)
+{
+	int rc;
+	struct vmlogrdr_priv_t * priv = filp->private_data;
+
+	while (priv->buffer_free) {
+		rc = vmlogrdr_receive_data(priv);
+		if (rc) {
+			rc = wait_event_interruptible(read_wait_queue,
+					atomic_read(&priv->receive_ready));
+			if (rc)
+				return rc;
+		}
+	}
+	/* copy only up to end of record */
+	if (count > priv->remaining)
+		count = priv->remaining;
+
+	if (copy_to_user(data, priv->current_position, count))
+		return -EFAULT;
+
+	*ppos += count;
+	priv->current_position += count;
+	priv->remaining -= count;
+
+	/* if all data has been transferred, set buffer free */
+	if (priv->remaining == 0)
+		priv->buffer_free = 1;
+
+	return count;
+}
+
+static ssize_t vmlogrdr_autopurge_store(struct device * dev,
+					struct device_attribute *attr,
+					const char * buf, size_t count)
+{
+	struct vmlogrdr_priv_t *priv = dev_get_drvdata(dev);
+	ssize_t ret = count;
+
+	switch (buf[0]) {
+	case '0':
+		priv->autopurge=0;
+		break;
+	case '1':
+		priv->autopurge=1;
+		break;
+	default:
+		ret = -EINVAL;
+	}
+	return ret;
+}
+
+
+static ssize_t vmlogrdr_autopurge_show(struct device *dev,
+				       struct device_attribute *attr,
+				       char *buf)
+{
+	struct vmlogrdr_priv_t *priv = dev_get_drvdata(dev);
+	return sprintf(buf, "%u\n", priv->autopurge);
+}
+
+
+static DEVICE_ATTR(autopurge, 0644, vmlogrdr_autopurge_show,
+		   vmlogrdr_autopurge_store);
+
+
+static ssize_t vmlogrdr_purge_store(struct device * dev,
+				    struct device_attribute *attr,
+				    const char * buf, size_t count)
+{
+
+	char cp_command[80];
+	char cp_response[80];
+	struct vmlogrdr_priv_t *priv = dev_get_drvdata(dev);
+
+	if (buf[0] != '1')
+		return -EINVAL;
+
+	memset(cp_command, 0x00, sizeof(cp_command));
+	memset(cp_response, 0x00, sizeof(cp_response));
+
+        /*
+	 * The recording command needs to be called with option QID
+	 * for guests that have previlege classes A or B.
+	 * Other guests will not recognize the command and we have to
+	 * issue the same command without the QID parameter.
+	 */
+
+	if (recording_class_AB)
+		snprintf(cp_command, sizeof(cp_command),
+			 "RECORDING %s PURGE QID * ",
+			 priv->recording_name);
+	else
+		snprintf(cp_command, sizeof(cp_command),
+			 "RECORDING %s PURGE ",
+			 priv->recording_name);
+
+	cpcmd(cp_command, cp_response, sizeof(cp_response), NULL);
+
+	return count;
+}
+
+
+static DEVICE_ATTR(purge, 0200, NULL, vmlogrdr_purge_store);
+
+
+static ssize_t vmlogrdr_autorecording_store(struct device *dev,
+					    struct device_attribute *attr,
+					    const char *buf, size_t count)
+{
+	struct vmlogrdr_priv_t *priv = dev_get_drvdata(dev);
+	ssize_t ret = count;
+
+	switch (buf[0]) {
+	case '0':
+		priv->autorecording=0;
+		break;
+	case '1':
+		priv->autorecording=1;
+		break;
+	default:
+		ret = -EINVAL;
+	}
+	return ret;
+}
+
+
+static ssize_t vmlogrdr_autorecording_show(struct device *dev,
+					   struct device_attribute *attr,
+					   char *buf)
+{
+	struct vmlogrdr_priv_t *priv = dev_get_drvdata(dev);
+	return sprintf(buf, "%u\n", priv->autorecording);
+}
+
+
+static DEVICE_ATTR(autorecording, 0644, vmlogrdr_autorecording_show,
+		   vmlogrdr_autorecording_store);
+
+
+static ssize_t vmlogrdr_recording_store(struct device * dev,
+					struct device_attribute *attr,
+					const char * buf, size_t count)
+{
+	struct vmlogrdr_priv_t *priv = dev_get_drvdata(dev);
+	ssize_t ret;
+
+	switch (buf[0]) {
+	case '0':
+		ret = vmlogrdr_recording(priv,0,0);
+		break;
+	case '1':
+		ret = vmlogrdr_recording(priv,1,0);
+		break;
+	default:
+		ret = -EINVAL;
+	}
+	if (ret)
+		return ret;
+	else
+		return count;
+
+}
+
+
+static DEVICE_ATTR(recording, 0200, NULL, vmlogrdr_recording_store);
+
+
+static ssize_t vmlogrdr_recording_status_show(struct device_driver *driver,
+					      char *buf)
+{
+
+	char cp_command[] = "QUERY RECORDING ";
+	int len;
+
+	cpcmd(cp_command, buf, 4096, NULL);
+	len = strlen(buf);
+	return len;
+}
+
+
+static DRIVER_ATTR(recording_status, 0444, vmlogrdr_recording_status_show,
+		   NULL);
+
+static struct attribute *vmlogrdr_attrs[] = {
+	&dev_attr_autopurge.attr,
+	&dev_attr_purge.attr,
+	&dev_attr_autorecording.attr,
+	&dev_attr_recording.attr,
+	NULL,
+};
+
+static int vmlogrdr_pm_prepare(struct device *dev)
+{
+	int rc;
+	struct vmlogrdr_priv_t *priv = dev_get_drvdata(dev);
+
+	rc = 0;
+	if (priv) {
+		spin_lock_bh(&priv->priv_lock);
+		if (priv->dev_in_use)
+			rc = -EBUSY;
+		spin_unlock_bh(&priv->priv_lock);
+	}
+	if (rc)
+		pr_err("vmlogrdr: device %s is busy. Refuse to suspend.\n",
+		       dev_name(dev));
+	return rc;
+}
+
+
+static struct dev_pm_ops vmlogrdr_pm_ops = {
+	.prepare = vmlogrdr_pm_prepare,
+};
+
+static struct attribute_group vmlogrdr_attr_group = {
+	.attrs = vmlogrdr_attrs,
+};
+
+static struct class *vmlogrdr_class;
+static struct device_driver vmlogrdr_driver = {
+	.name = "vmlogrdr",
+	.bus  = &iucv_bus,
+	.pm = &vmlogrdr_pm_ops,
+};
+
+
+static int vmlogrdr_register_driver(void)
+{
+	int ret;
+
+	/* Register with iucv driver */
+	ret = iucv_register(&vmlogrdr_iucv_handler, 1);
+	if (ret)
+		goto out;
+
+	ret = driver_register(&vmlogrdr_driver);
+	if (ret)
+		goto out_iucv;
+
+	ret = driver_create_file(&vmlogrdr_driver,
+				 &driver_attr_recording_status);
+	if (ret)
+		goto out_driver;
+
+	vmlogrdr_class = class_create(THIS_MODULE, "vmlogrdr");
+	if (IS_ERR(vmlogrdr_class)) {
+		ret = PTR_ERR(vmlogrdr_class);
+		vmlogrdr_class = NULL;
+		goto out_attr;
+	}
+	return 0;
+
+out_attr:
+	driver_remove_file(&vmlogrdr_driver, &driver_attr_recording_status);
+out_driver:
+	driver_unregister(&vmlogrdr_driver);
+out_iucv:
+	iucv_unregister(&vmlogrdr_iucv_handler, 1);
+out:
+	return ret;
+}
+
+
+static void vmlogrdr_unregister_driver(void)
+{
+	class_destroy(vmlogrdr_class);
+	vmlogrdr_class = NULL;
+	driver_remove_file(&vmlogrdr_driver, &driver_attr_recording_status);
+	driver_unregister(&vmlogrdr_driver);
+	iucv_unregister(&vmlogrdr_iucv_handler, 1);
+}
+
+
+static int vmlogrdr_register_device(struct vmlogrdr_priv_t *priv)
+{
+	struct device *dev;
+	int ret;
+
+	dev = kzalloc(sizeof(struct device), GFP_KERNEL);
+	if (dev) {
+		dev_set_name(dev, priv->internal_name);
+		dev->bus = &iucv_bus;
+		dev->parent = iucv_root;
+		dev->driver = &vmlogrdr_driver;
+		dev_set_drvdata(dev, priv);
+		/*
+		 * The release function could be called after the
+		 * module has been unloaded. It's _only_ task is to
+		 * free the struct. Therefore, we specify kfree()
+		 * directly here. (Probably a little bit obfuscating
+		 * but legitime ...).
+		 */
+		dev->release = (void (*)(struct device *))kfree;
+	} else
+		return -ENOMEM;
+	ret = device_register(dev);
+	if (ret) {
+		put_device(dev);
+		return ret;
+	}
+
+	ret = sysfs_create_group(&dev->kobj, &vmlogrdr_attr_group);
+	if (ret) {
+		device_unregister(dev);
+		return ret;
+	}
+	priv->class_device = device_create(vmlogrdr_class, dev,
+					   MKDEV(vmlogrdr_major,
+						 priv->minor_num),
+					   priv, "%s", dev_name(dev));
+	if (IS_ERR(priv->class_device)) {
+		ret = PTR_ERR(priv->class_device);
+		priv->class_device=NULL;
+		sysfs_remove_group(&dev->kobj, &vmlogrdr_attr_group);
+		device_unregister(dev);
+		return ret;
+	}
+	priv->device = dev;
+	return 0;
+}
+
+
+static int vmlogrdr_unregister_device(struct vmlogrdr_priv_t *priv)
+{
+	device_destroy(vmlogrdr_class, MKDEV(vmlogrdr_major, priv->minor_num));
+	if (priv->device != NULL) {
+		sysfs_remove_group(&priv->device->kobj, &vmlogrdr_attr_group);
+		device_unregister(priv->device);
+		priv->device=NULL;
+	}
+	return 0;
+}
+
+
+static int vmlogrdr_register_cdev(dev_t dev)
+{
+	int rc = 0;
+	vmlogrdr_cdev = cdev_alloc();
+	if (!vmlogrdr_cdev) {
+		return -ENOMEM;
+	}
+	vmlogrdr_cdev->owner = THIS_MODULE;
+	vmlogrdr_cdev->ops = &vmlogrdr_fops;
+	vmlogrdr_cdev->dev = dev;
+	rc = cdev_add(vmlogrdr_cdev, vmlogrdr_cdev->dev, MAXMINOR);
+	if (!rc)
+		return 0;
+
+	// cleanup: cdev is not fully registered, no cdev_del here!
+	kobject_put(&vmlogrdr_cdev->kobj);
+	vmlogrdr_cdev=NULL;
+	return rc;
+}
+
+
+static void vmlogrdr_cleanup(void)
+{
+        int i;
+
+	if (vmlogrdr_cdev) {
+		cdev_del(vmlogrdr_cdev);
+		vmlogrdr_cdev=NULL;
+	}
+	for (i=0; i < MAXMINOR; ++i ) {
+		vmlogrdr_unregister_device(&sys_ser[i]);
+		free_page((unsigned long)sys_ser[i].buffer);
+	}
+	vmlogrdr_unregister_driver();
+	if (vmlogrdr_major) {
+		unregister_chrdev_region(MKDEV(vmlogrdr_major, 0), MAXMINOR);
+		vmlogrdr_major=0;
+	}
+}
+
+
+static int __init vmlogrdr_init(void)
+{
+	int rc;
+	int i;
+	dev_t dev;
+
+	if (! MACHINE_IS_VM) {
+		pr_err("not running under VM, driver not loaded.\n");
+		return -ENODEV;
+	}
+
+        recording_class_AB = vmlogrdr_get_recording_class_AB();
+
+	rc = alloc_chrdev_region(&dev, 0, MAXMINOR, "vmlogrdr");
+	if (rc)
+		return rc;
+	vmlogrdr_major = MAJOR(dev);
+
+	rc=vmlogrdr_register_driver();
+	if (rc)
+		goto cleanup;
+
+	for (i=0; i < MAXMINOR; ++i ) {
+		sys_ser[i].buffer = (char *) get_zeroed_page(GFP_KERNEL);
+		if (!sys_ser[i].buffer) {
+			rc = -ENOMEM;
+			break;
+		}
+		sys_ser[i].current_position = sys_ser[i].buffer;
+		rc=vmlogrdr_register_device(&sys_ser[i]);
+		if (rc)
+			break;
+	}
+	if (rc)
+		goto cleanup;
+
+	rc = vmlogrdr_register_cdev(dev);
+	if (rc)
+		goto cleanup;
+	return 0;
+
+cleanup:
+	vmlogrdr_cleanup();
+	return rc;
+}
+
+
+static void __exit vmlogrdr_exit(void)
+{
+	vmlogrdr_cleanup();
+	return;
+}
+
+
+module_init(vmlogrdr_init);
+module_exit(vmlogrdr_exit);

kernel/drivers/s390/char/vmur.h

@@ -0,0 +1,110 @@
+/*
+ * Linux driver for System z and s390 unit record devices
+ * (z/VM virtual punch, reader, printer)
+ *
+ * Copyright IBM Corp. 2001, 2007
+ * Authors: Malcolm Beattie <beattiem@uk.ibm.com>
+ *	    Michael Holzheu <holzheu@de.ibm.com>
+ *	    Frank Munzert <munzert@de.ibm.com>
+ */
+
+#ifndef _VMUR_H_
+#define _VMUR_H_
+
+#define DEV_CLASS_UR_I 0x20 /* diag210 unit record input device class */
+#define DEV_CLASS_UR_O 0x10 /* diag210 unit record output device class */
+/*
+ * we only support z/VM's default unit record devices:
+ * both in SPOOL directory control statement and in CP DEFINE statement
+ *	RDR defaults to 2540 reader
+ *	PUN defaults to 2540 punch
+ *	PRT defaults to 1403 printer
+ */
+#define READER_PUNCH_DEVTYPE	0x2540
+#define PRINTER_DEVTYPE		0x1403
+
+/* z/VM spool file control block SFBLOK */
+struct file_control_block {
+	char reserved_1[8];
+	char user_owner[8];
+	char user_orig[8];
+	__s32 data_recs;
+	__s16 rec_len;
+	__s16 file_num;
+	__u8  file_stat;
+	__u8  dev_type;
+	char  reserved_2[6];
+	char  file_name[12];
+	char  file_type[12];
+	char  create_date[8];
+	char  create_time[8];
+	char  reserved_3[6];
+	__u8  file_class;
+	__u8  sfb_lok;
+	__u64 distr_code;
+	__u32 reserved_4;
+	__u8  current_starting_copy_number;
+	__u8  sfblock_cntrl_flags;
+	__u8  reserved_5;
+	__u8  more_status_flags;
+	char  rest[200];
+} __attribute__ ((packed));
+
+#define FLG_SYSTEM_HOLD	0x04
+#define FLG_CP_DUMP	0x10
+#define FLG_USER_HOLD	0x20
+#define FLG_IN_USE	0x80
+
+/*
+ * A struct urdev is created for each ur device that is made available
+ * via the ccw_device driver model.
+ */
+struct urdev {
+	struct ccw_device *cdev;	/* Backpointer to ccw device */
+	struct mutex io_mutex;		/* Serialises device IO */
+	struct completion *io_done;	/* do_ur_io waits; irq completes */
+	struct device *device;
+	struct cdev *char_device;
+	struct ccw_dev_id dev_id;	/* device id */
+	size_t reclen;			/* Record length for *write* CCWs */
+	int class;			/* VM device class */
+	int io_request_rc;		/* return code from I/O request */
+	atomic_t ref_count;		/* reference counter */
+	wait_queue_head_t wait;		/* wait queue to serialize open */
+	int open_flag;			/* "urdev is open" flag */
+	spinlock_t open_lock;		/* serialize critical sections */
+};
+
+/*
+ * A struct urfile is allocated at open() time for each device and
+ * freed on release().
+ */
+struct urfile {
+	struct urdev *urd;
+	unsigned int flags;
+	size_t dev_reclen;
+	__u16 file_reclen;
+};
+
+/*
+ * Device major/minor definitions.
+ */
+
+#define UR_MAJOR 0	/* get dynamic major */
+/*
+ * We map minor numbers directly to device numbers (0-FFFF) for simplicity.
+ * This avoids having to allocate (and manage) slot numbers.
+ */
+#define NUM_MINORS 65536
+
+/* Limiting each I/O to 511 records limits chan prog to 4KB (511 r/w + 1 NOP) */
+#define MAX_RECS_PER_IO		511
+#define WRITE_CCW_CMD		0x01
+
+#define TRACE(x...) debug_sprintf_event(vmur_dbf, 1, x)
+#define CCWDEV_CU_DI(cutype, di) \
+		CCW_DEVICE(cutype, 0x00), .driver_info = (di)
+
+#define FILE_RECLEN_OFFSET	4064 /* reclen offset in spool data block */
+
+#endif /* _VMUR_H_ */

kernel/drivers/s390/char/vmwatchdog.c

@@ -0,0 +1,326 @@
+/*
+ * Watchdog implementation based on z/VM Watchdog Timer API
+ *
+ * Copyright IBM Corp. 2004,2009
+ *
+ * The user space watchdog daemon can use this driver as
+ * /dev/vmwatchdog to have z/VM execute the specified CP
+ * command when the timeout expires. The default command is
+ * "IPL", which which cause an immediate reboot.
+ */
+#define KMSG_COMPONENT "vmwatchdog"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/kernel.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/suspend.h>
+#include <linux/watchdog.h>
+#include <linux/smp_lock.h>
+
+#include <asm/ebcdic.h>
+#include <asm/io.h>
+#include <asm/uaccess.h>
+
+#define MAX_CMDLEN 240
+#define MIN_INTERVAL 15
+static char vmwdt_cmd[MAX_CMDLEN] = "IPL";
+static int vmwdt_conceal;
+
+static int vmwdt_nowayout = WATCHDOG_NOWAYOUT;
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Arnd Bergmann <arndb@de.ibm.com>");
+MODULE_DESCRIPTION("z/VM Watchdog Timer");
+module_param_string(cmd, vmwdt_cmd, MAX_CMDLEN, 0644);
+MODULE_PARM_DESC(cmd, "CP command that is run when the watchdog triggers");
+module_param_named(conceal, vmwdt_conceal, bool, 0644);
+MODULE_PARM_DESC(conceal, "Enable the CONCEAL CP option while the watchdog "
+		" is active");
+module_param_named(nowayout, vmwdt_nowayout, bool, 0);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started"
+		" (default=CONFIG_WATCHDOG_NOWAYOUT)");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
+
+static unsigned int vmwdt_interval = 60;
+static unsigned long vmwdt_is_open;
+static int vmwdt_expect_close;
+
+#define VMWDT_OPEN	0	/* devnode is open or suspend in progress */
+#define VMWDT_RUNNING	1	/* The watchdog is armed */
+
+enum vmwdt_func {
+	/* function codes */
+	wdt_init   = 0,
+	wdt_change = 1,
+	wdt_cancel = 2,
+	/* flags */
+	wdt_conceal = 0x80000000,
+};
+
+static int __diag288(enum vmwdt_func func, unsigned int timeout,
+			    char *cmd, size_t len)
+{
+	register unsigned long __func asm("2") = func;
+	register unsigned long __timeout asm("3") = timeout;
+	register unsigned long __cmdp asm("4") = virt_to_phys(cmd);
+	register unsigned long __cmdl asm("5") = len;
+	int err;
+
+	err = -EINVAL;
+	asm volatile(
+		"	diag	%1,%3,0x288\n"
+		"0:	la	%0,0\n"
+		"1:\n"
+		EX_TABLE(0b,1b)
+		: "+d" (err) : "d"(__func), "d"(__timeout),
+		  "d"(__cmdp), "d"(__cmdl) : "1", "cc");
+	return err;
+}
+
+static int vmwdt_keepalive(void)
+{
+	/* we allocate new memory every time to avoid having
+	 * to track the state. static allocation is not an
+	 * option since that might not be contiguous in real
+	 * storage in case of a modular build */
+	static char *ebc_cmd;
+	size_t len;
+	int ret;
+	unsigned int func;
+
+	ebc_cmd = kmalloc(MAX_CMDLEN, GFP_KERNEL);
+	if (!ebc_cmd)
+		return -ENOMEM;
+
+	len = strlcpy(ebc_cmd, vmwdt_cmd, MAX_CMDLEN);
+	ASCEBC(ebc_cmd, MAX_CMDLEN);
+	EBC_TOUPPER(ebc_cmd, MAX_CMDLEN);
+
+	func = vmwdt_conceal ? (wdt_init | wdt_conceal) : wdt_init;
+	set_bit(VMWDT_RUNNING, &vmwdt_is_open);
+	ret = __diag288(func, vmwdt_interval, ebc_cmd, len);
+	WARN_ON(ret != 0);
+	kfree(ebc_cmd);
+	return ret;
+}
+
+static int vmwdt_disable(void)
+{
+	int ret = __diag288(wdt_cancel, 0, "", 0);
+	WARN_ON(ret != 0);
+	clear_bit(VMWDT_RUNNING, &vmwdt_is_open);
+	return ret;
+}
+
+static int __init vmwdt_probe(void)
+{
+	/* there is no real way to see if the watchdog is supported,
+	 * so we try initializing it with a NOP command ("BEGIN")
+	 * that won't cause any harm even if the following disable
+	 * fails for some reason */
+	static char __initdata ebc_begin[] = {
+		194, 197, 199, 201, 213
+	};
+	if (__diag288(wdt_init, 15, ebc_begin, sizeof(ebc_begin)) != 0)
+		return -EINVAL;
+	return vmwdt_disable();
+}
+
+static int vmwdt_open(struct inode *i, struct file *f)
+{
+	int ret;
+	lock_kernel();
+	if (test_and_set_bit(VMWDT_OPEN, &vmwdt_is_open)) {
+		unlock_kernel();
+		return -EBUSY;
+	}
+	ret = vmwdt_keepalive();
+	if (ret)
+		clear_bit(VMWDT_OPEN, &vmwdt_is_open);
+	unlock_kernel();
+	return ret ? ret : nonseekable_open(i, f);
+}
+
+static int vmwdt_close(struct inode *i, struct file *f)
+{
+	if (vmwdt_expect_close == 42)
+		vmwdt_disable();
+	vmwdt_expect_close = 0;
+	clear_bit(VMWDT_OPEN, &vmwdt_is_open);
+	return 0;
+}
+
+static struct watchdog_info vmwdt_info = {
+	.options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE,
+	.firmware_version = 0,
+	.identity = "z/VM Watchdog Timer",
+};
+
+static int vmwdt_ioctl(struct inode *i, struct file *f,
+			  unsigned int cmd, unsigned long arg)
+{
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		if (copy_to_user((void __user *)arg, &vmwdt_info,
+					sizeof(vmwdt_info)))
+			return -EFAULT;
+		return 0;
+	case WDIOC_GETSTATUS:
+	case WDIOC_GETBOOTSTATUS:
+		return put_user(0, (int __user *)arg);
+	case WDIOC_GETTEMP:
+		return -EINVAL;
+	case WDIOC_SETOPTIONS:
+		{
+			int options, ret;
+			if (get_user(options, (int __user *)arg))
+				return -EFAULT;
+			ret = -EINVAL;
+			if (options & WDIOS_DISABLECARD) {
+				ret = vmwdt_disable();
+				if (ret)
+					return ret;
+			}
+			if (options & WDIOS_ENABLECARD) {
+				ret = vmwdt_keepalive();
+			}
+			return ret;
+		}
+	case WDIOC_GETTIMEOUT:
+		return put_user(vmwdt_interval, (int __user *)arg);
+	case WDIOC_SETTIMEOUT:
+		{
+			int interval;
+			if (get_user(interval, (int __user *)arg))
+				return -EFAULT;
+			if (interval < MIN_INTERVAL)
+				return -EINVAL;
+			vmwdt_interval = interval;
+		}
+		return vmwdt_keepalive();
+	case WDIOC_KEEPALIVE:
+		return vmwdt_keepalive();
+	}
+
+	return -EINVAL;
+}
+
+static ssize_t vmwdt_write(struct file *f, const char __user *buf,
+				size_t count, loff_t *ppos)
+{
+	if(count) {
+		if (!vmwdt_nowayout) {
+			size_t i;
+
+			/* note: just in case someone wrote the magic character
+			 * five months ago... */
+			vmwdt_expect_close = 0;
+
+			for (i = 0; i != count; i++) {
+				char c;
+				if (get_user(c, buf+i))
+					return -EFAULT;
+				if (c == 'V')
+					vmwdt_expect_close = 42;
+			}
+		}
+		/* someone wrote to us, we should restart timer */
+		vmwdt_keepalive();
+	}
+	return count;
+}
+
+static int vmwdt_resume(void)
+{
+	clear_bit(VMWDT_OPEN, &vmwdt_is_open);
+	return NOTIFY_DONE;
+}
+
+/*
+ * It makes no sense to go into suspend while the watchdog is running.
+ * Depending on the memory size, the watchdog might trigger, while we
+ * are still saving the memory.
+ * We reuse the open flag to ensure that suspend and watchdog open are
+ * exclusive operations
+ */
+static int vmwdt_suspend(void)
+{
+	if (test_and_set_bit(VMWDT_OPEN, &vmwdt_is_open)) {
+		pr_err("The system cannot be suspended while the watchdog"
+			" is in use\n");
+		return NOTIFY_BAD;
+	}
+	if (test_bit(VMWDT_RUNNING, &vmwdt_is_open)) {
+		clear_bit(VMWDT_OPEN, &vmwdt_is_open);
+		pr_err("The system cannot be suspended while the watchdog"
+			" is running\n");
+		return NOTIFY_BAD;
+	}
+	return NOTIFY_DONE;
+}
+
+/*
+ * This function is called for suspend and resume.
+ */
+static int vmwdt_power_event(struct notifier_block *this, unsigned long event,
+			     void *ptr)
+{
+	switch (event) {
+	case PM_POST_HIBERNATION:
+	case PM_POST_SUSPEND:
+		return vmwdt_resume();
+	case PM_HIBERNATION_PREPARE:
+	case PM_SUSPEND_PREPARE:
+		return vmwdt_suspend();
+	default:
+		return NOTIFY_DONE;
+	}
+}
+
+static struct notifier_block vmwdt_power_notifier = {
+	.notifier_call = vmwdt_power_event,
+};
+
+static const struct file_operations vmwdt_fops = {
+	.open    = &vmwdt_open,
+	.release = &vmwdt_close,
+	.ioctl   = &vmwdt_ioctl,
+	.write   = &vmwdt_write,
+	.owner   = THIS_MODULE,
+};
+
+static struct miscdevice vmwdt_dev = {
+	.minor      = WATCHDOG_MINOR,
+	.name       = "watchdog",
+	.fops       = &vmwdt_fops,
+};
+
+static int __init vmwdt_init(void)
+{
+	int ret;
+
+	ret = vmwdt_probe();
+	if (ret)
+		return ret;
+	ret = register_pm_notifier(&vmwdt_power_notifier);
+	if (ret)
+		return ret;
+	ret = misc_register(&vmwdt_dev);
+	if (ret) {
+		unregister_pm_notifier(&vmwdt_power_notifier);
+		return ret;
+	}
+	return 0;
+}
+module_init(vmwdt_init);
+
+static void __exit vmwdt_exit(void)
+{
+	unregister_pm_notifier(&vmwdt_power_notifier);
+	misc_deregister(&vmwdt_dev);
+}
+module_exit(vmwdt_exit);

kernel/drivers/s390/char/zcore.c

@@ -0,0 +1,819 @@
+/*
+ * zcore module to export memory content and register sets for creating system
+ * dumps on SCSI disks (zfcpdump). The "zcore/mem" debugfs file shows the same
+ * dump format as s390 standalone dumps.
+ *
+ * For more information please refer to Documentation/s390/zfcpdump.txt
+ *
+ * Copyright IBM Corp. 2003,2008
+ * Author(s): Michael Holzheu
+ */
+
+#define KMSG_COMPONENT "zdump"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/init.h>
+#include <linux/miscdevice.h>
+#include <linux/debugfs.h>
+#include <asm/ipl.h>
+#include <asm/sclp.h>
+#include <asm/setup.h>
+#include <asm/sigp.h>
+#include <asm/uaccess.h>
+#include <asm/debug.h>
+#include <asm/processor.h>
+#include <asm/irqflags.h>
+#include <asm/checksum.h>
+#include "sclp.h"
+
+#define TRACE(x...) debug_sprintf_event(zcore_dbf, 1, x)
+
+#define TO_USER		0
+#define TO_KERNEL	1
+#define CHUNK_INFO_SIZE	34 /* 2 16-byte char, each followed by blank */
+
+enum arch_id {
+	ARCH_S390	= 0,
+	ARCH_S390X	= 1,
+};
+
+/* dump system info */
+
+struct sys_info {
+	enum arch_id	arch;
+	unsigned long	sa_base;
+	u32		sa_size;
+	int		cpu_map[NR_CPUS];
+	unsigned long	mem_size;
+	union save_area	lc_mask;
+};
+
+struct ipib_info {
+	unsigned long	ipib;
+	u32		checksum;
+}  __attribute__((packed));
+
+static struct sys_info sys_info;
+static struct debug_info *zcore_dbf;
+static int hsa_available;
+static struct dentry *zcore_dir;
+static struct dentry *zcore_file;
+static struct dentry *zcore_memmap_file;
+static struct dentry *zcore_reipl_file;
+static struct ipl_parameter_block *ipl_block;
+
+/*
+ * Copy memory from HSA to kernel or user memory (not reentrant):
+ *
+ * @dest:  Kernel or user buffer where memory should be copied to
+ * @src:   Start address within HSA where data should be copied
+ * @count: Size of buffer, which should be copied
+ * @mode:  Either TO_KERNEL or TO_USER
+ */
+static int memcpy_hsa(void *dest, unsigned long src, size_t count, int mode)
+{
+	int offs, blk_num;
+	static char buf[PAGE_SIZE] __attribute__((__aligned__(PAGE_SIZE)));
+
+	if (count == 0)
+		return 0;
+
+	/* copy first block */
+	offs = 0;
+	if ((src % PAGE_SIZE) != 0) {
+		blk_num = src / PAGE_SIZE + 2;
+		if (sclp_sdias_copy(buf, blk_num, 1)) {
+			TRACE("sclp_sdias_copy() failed\n");
+			return -EIO;
+		}
+		offs = min((PAGE_SIZE - (src % PAGE_SIZE)), count);
+		if (mode == TO_USER) {
+			if (copy_to_user((__force __user void*) dest,
+					 buf + (src % PAGE_SIZE), offs))
+				return -EFAULT;
+		} else
+			memcpy(dest, buf + (src % PAGE_SIZE), offs);
+	}
+	if (offs == count)
+		goto out;
+
+	/* copy middle */
+	for (; (offs + PAGE_SIZE) <= count; offs += PAGE_SIZE) {
+		blk_num = (src + offs) / PAGE_SIZE + 2;
+		if (sclp_sdias_copy(buf, blk_num, 1)) {
+			TRACE("sclp_sdias_copy() failed\n");
+			return -EIO;
+		}
+		if (mode == TO_USER) {
+			if (copy_to_user((__force __user void*) dest + offs,
+					 buf, PAGE_SIZE))
+				return -EFAULT;
+		} else
+			memcpy(dest + offs, buf, PAGE_SIZE);
+	}
+	if (offs == count)
+		goto out;
+
+	/* copy last block */
+	blk_num = (src + offs) / PAGE_SIZE + 2;
+	if (sclp_sdias_copy(buf, blk_num, 1)) {
+		TRACE("sclp_sdias_copy() failed\n");
+		return -EIO;
+	}
+	if (mode == TO_USER) {
+		if (copy_to_user((__force __user void*) dest + offs, buf,
+				 PAGE_SIZE))
+			return -EFAULT;
+	} else
+		memcpy(dest + offs, buf, count - offs);
+out:
+	return 0;
+}
+
+static int memcpy_hsa_user(void __user *dest, unsigned long src, size_t count)
+{
+	return memcpy_hsa((void __force *) dest, src, count, TO_USER);
+}
+
+static int memcpy_hsa_kernel(void *dest, unsigned long src, size_t count)
+{
+	return memcpy_hsa(dest, src, count, TO_KERNEL);
+}
+
+static int memcpy_real(void *dest, unsigned long src, size_t count)
+{
+	unsigned long flags;
+	int rc = -EFAULT;
+	register unsigned long _dest asm("2") = (unsigned long) dest;
+	register unsigned long _len1 asm("3") = (unsigned long) count;
+	register unsigned long _src  asm("4") = src;
+	register unsigned long _len2 asm("5") = (unsigned long) count;
+
+	if (count == 0)
+		return 0;
+	flags = __raw_local_irq_stnsm(0xf8UL); /* switch to real mode */
+	asm volatile (
+		"0:	mvcle	%1,%2,0x0\n"
+		"1:	jo	0b\n"
+		"	lhi	%0,0x0\n"
+		"2:\n"
+		EX_TABLE(1b,2b)
+		: "+d" (rc), "+d" (_dest), "+d" (_src), "+d" (_len1),
+		  "+d" (_len2), "=m" (*((long*)dest))
+		: "m" (*((long*)src))
+		: "cc", "memory");
+	__raw_local_irq_ssm(flags);
+
+	return rc;
+}
+
+static int memcpy_real_user(void __user *dest, unsigned long src, size_t count)
+{
+	static char buf[4096];
+	int offs = 0, size;
+
+	while (offs < count) {
+		size = min(sizeof(buf), count - offs);
+		if (memcpy_real(buf, src + offs, size))
+			return -EFAULT;
+		if (copy_to_user(dest + offs, buf, size))
+			return -EFAULT;
+		offs += size;
+	}
+	return 0;
+}
+
+#ifdef __s390x__
+/*
+ * Convert s390x (64 bit) cpu info to s390 (32 bit) cpu info
+ */
+static void __init s390x_to_s390_regs(union save_area *out, union save_area *in,
+				      int cpu)
+{
+	int i;
+
+	for (i = 0; i < 16; i++) {
+		out->s390.gp_regs[i] = in->s390x.gp_regs[i] & 0x00000000ffffffff;
+		out->s390.acc_regs[i] = in->s390x.acc_regs[i];
+		out->s390.ctrl_regs[i] =
+			in->s390x.ctrl_regs[i] & 0x00000000ffffffff;
+	}
+	/* locore for 31 bit has only space for fpregs 0,2,4,6 */
+	out->s390.fp_regs[0] = in->s390x.fp_regs[0];
+	out->s390.fp_regs[1] = in->s390x.fp_regs[2];
+	out->s390.fp_regs[2] = in->s390x.fp_regs[4];
+	out->s390.fp_regs[3] = in->s390x.fp_regs[6];
+	memcpy(&(out->s390.psw[0]), &(in->s390x.psw[0]), 4);
+	out->s390.psw[1] |= 0x8; /* set bit 12 */
+	memcpy(&(out->s390.psw[4]),&(in->s390x.psw[12]), 4);
+	out->s390.psw[4] |= 0x80; /* set (31bit) addressing bit */
+	out->s390.pref_reg = in->s390x.pref_reg;
+	out->s390.timer = in->s390x.timer;
+	out->s390.clk_cmp = in->s390x.clk_cmp;
+}
+
+static void __init s390x_to_s390_save_areas(void)
+{
+	int i = 1;
+	static union save_area tmp;
+
+	while (zfcpdump_save_areas[i]) {
+		s390x_to_s390_regs(&tmp, zfcpdump_save_areas[i], i);
+		memcpy(zfcpdump_save_areas[i], &tmp, sizeof(tmp));
+		i++;
+	}
+}
+
+#endif /* __s390x__ */
+
+static int __init init_cpu_info(enum arch_id arch)
+{
+	union save_area *sa;
+
+	/* get info for boot cpu from lowcore, stored in the HSA */
+
+	sa = kmalloc(sizeof(*sa), GFP_KERNEL);
+	if (!sa)
+		return -ENOMEM;
+	if (memcpy_hsa_kernel(sa, sys_info.sa_base, sys_info.sa_size) < 0) {
+		TRACE("could not copy from HSA\n");
+		kfree(sa);
+		return -EIO;
+	}
+	zfcpdump_save_areas[0] = sa;
+
+#ifdef __s390x__
+	/* convert s390x regs to s390, if we are dumping an s390 Linux */
+
+	if (arch == ARCH_S390)
+		s390x_to_s390_save_areas();
+#endif
+
+	return 0;
+}
+
+static DEFINE_MUTEX(zcore_mutex);
+
+#define DUMP_VERSION	0x3
+#define DUMP_MAGIC	0xa8190173618f23fdULL
+#define DUMP_ARCH_S390X	2
+#define DUMP_ARCH_S390	1
+#define HEADER_SIZE	4096
+
+/* dump header dumped according to s390 crash dump format */
+
+struct zcore_header {
+	u64 magic;
+	u32 version;
+	u32 header_size;
+	u32 dump_level;
+	u32 page_size;
+	u64 mem_size;
+	u64 mem_start;
+	u64 mem_end;
+	u32 num_pages;
+	u32 pad1;
+	u64 tod;
+	struct cpuid cpu_id;
+	u32 arch_id;
+	u32 volnr;
+	u32 build_arch;
+	u64 rmem_size;
+	char pad2[4016];
+} __attribute__((packed,__aligned__(16)));
+
+static struct zcore_header zcore_header = {
+	.magic		= DUMP_MAGIC,
+	.version	= DUMP_VERSION,
+	.header_size	= 4096,
+	.dump_level	= 0,
+	.page_size	= PAGE_SIZE,
+	.mem_start	= 0,
+#ifdef __s390x__
+	.build_arch	= DUMP_ARCH_S390X,
+#else
+	.build_arch	= DUMP_ARCH_S390,
+#endif
+};
+
+/*
+ * Copy lowcore info to buffer. Use map in order to copy only register parts.
+ *
+ * @buf:    User buffer
+ * @sa:     Pointer to save area
+ * @sa_off: Offset in save area to copy
+ * @len:    Number of bytes to copy
+ */
+static int copy_lc(void __user *buf, void *sa, int sa_off, int len)
+{
+	int i;
+	char *lc_mask = (char*)&sys_info.lc_mask;
+
+	for (i = 0; i < len; i++) {
+		if (!lc_mask[i + sa_off])
+			continue;
+		if (copy_to_user(buf + i, sa + sa_off + i, 1))
+			return -EFAULT;
+	}
+	return 0;
+}
+
+/*
+ * Copy lowcores info to memory, if necessary
+ *
+ * @buf:   User buffer
+ * @addr:  Start address of buffer in dump memory
+ * @count: Size of buffer
+ */
+static int zcore_add_lc(char __user *buf, unsigned long start, size_t count)
+{
+	unsigned long end;
+	int i = 0;
+
+	if (count == 0)
+		return 0;
+
+	end = start + count;
+	while (zfcpdump_save_areas[i]) {
+		unsigned long cp_start, cp_end; /* copy range */
+		unsigned long sa_start, sa_end; /* save area range */
+		unsigned long prefix;
+		unsigned long sa_off, len, buf_off;
+
+		if (sys_info.arch == ARCH_S390)
+			prefix = zfcpdump_save_areas[i]->s390.pref_reg;
+		else
+			prefix = zfcpdump_save_areas[i]->s390x.pref_reg;
+
+		sa_start = prefix + sys_info.sa_base;
+		sa_end = prefix + sys_info.sa_base + sys_info.sa_size;
+
+		if ((end < sa_start) || (start > sa_end))
+			goto next;
+		cp_start = max(start, sa_start);
+		cp_end = min(end, sa_end);
+
+		buf_off = cp_start - start;
+		sa_off = cp_start - sa_start;
+		len = cp_end - cp_start;
+
+		TRACE("copy_lc for: %lx\n", start);
+		if (copy_lc(buf + buf_off, zfcpdump_save_areas[i], sa_off, len))
+			return -EFAULT;
+next:
+		i++;
+	}
+	return 0;
+}
+
+/*
+ * Read routine for zcore character device
+ * First 4K are dump header
+ * Next 32MB are HSA Memory
+ * Rest is read from absolute Memory
+ */
+static ssize_t zcore_read(struct file *file, char __user *buf, size_t count,
+			  loff_t *ppos)
+{
+	unsigned long mem_start; /* Start address in memory */
+	size_t mem_offs;	 /* Offset in dump memory */
+	size_t hdr_count;	 /* Size of header part of output buffer */
+	size_t size;
+	int rc;
+
+	mutex_lock(&zcore_mutex);
+
+	if (*ppos > (sys_info.mem_size + HEADER_SIZE)) {
+		rc = -EINVAL;
+		goto fail;
+	}
+
+	count = min(count, (size_t) (sys_info.mem_size + HEADER_SIZE - *ppos));
+
+	/* Copy dump header */
+	if (*ppos < HEADER_SIZE) {
+		size = min(count, (size_t) (HEADER_SIZE - *ppos));
+		if (copy_to_user(buf, &zcore_header + *ppos, size)) {
+			rc = -EFAULT;
+			goto fail;
+		}
+		hdr_count = size;
+		mem_start = 0;
+	} else {
+		hdr_count = 0;
+		mem_start = *ppos - HEADER_SIZE;
+	}
+
+	mem_offs = 0;
+
+	/* Copy from HSA data */
+	if (*ppos < (ZFCPDUMP_HSA_SIZE + HEADER_SIZE)) {
+		size = min((count - hdr_count), (size_t) (ZFCPDUMP_HSA_SIZE
+			   - mem_start));
+		rc = memcpy_hsa_user(buf + hdr_count, mem_start, size);
+		if (rc)
+			goto fail;
+
+		mem_offs += size;
+	}
+
+	/* Copy from real mem */
+	size = count - mem_offs - hdr_count;
+	rc = memcpy_real_user(buf + hdr_count + mem_offs, mem_start + mem_offs,
+			      size);
+	if (rc)
+		goto fail;
+
+	/*
+	 * Since s390 dump analysis tools like lcrash or crash
+	 * expect register sets in the prefix pages of the cpus,
+	 * we copy them into the read buffer, if necessary.
+	 * buf + hdr_count: Start of memory part of output buffer
+	 * mem_start: Start memory address to copy from
+	 * count - hdr_count: Size of memory area to copy
+	 */
+	if (zcore_add_lc(buf + hdr_count, mem_start, count - hdr_count)) {
+		rc = -EFAULT;
+		goto fail;
+	}
+	*ppos += count;
+fail:
+	mutex_unlock(&zcore_mutex);
+	return (rc < 0) ? rc : count;
+}
+
+static int zcore_open(struct inode *inode, struct file *filp)
+{
+	if (!hsa_available)
+		return -ENODATA;
+	else
+		return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
+}
+
+static int zcore_release(struct inode *inode, struct file *filep)
+{
+	diag308(DIAG308_REL_HSA, NULL);
+	hsa_available = 0;
+	return 0;
+}
+
+static loff_t zcore_lseek(struct file *file, loff_t offset, int orig)
+{
+	loff_t rc;
+
+	mutex_lock(&zcore_mutex);
+	switch (orig) {
+	case 0:
+		file->f_pos = offset;
+		rc = file->f_pos;
+		break;
+	case 1:
+		file->f_pos += offset;
+		rc = file->f_pos;
+		break;
+	default:
+		rc = -EINVAL;
+	}
+	mutex_unlock(&zcore_mutex);
+	return rc;
+}
+
+static const struct file_operations zcore_fops = {
+	.owner		= THIS_MODULE,
+	.llseek		= zcore_lseek,
+	.read		= zcore_read,
+	.open		= zcore_open,
+	.release	= zcore_release,
+};
+
+static ssize_t zcore_memmap_read(struct file *filp, char __user *buf,
+				 size_t count, loff_t *ppos)
+{
+	return simple_read_from_buffer(buf, count, ppos, filp->private_data,
+				       MEMORY_CHUNKS * CHUNK_INFO_SIZE);
+}
+
+static int zcore_memmap_open(struct inode *inode, struct file *filp)
+{
+	int i;
+	char *buf;
+	struct mem_chunk *chunk_array;
+
+	chunk_array = kzalloc(MEMORY_CHUNKS * sizeof(struct mem_chunk),
+			      GFP_KERNEL);
+	if (!chunk_array)
+		return -ENOMEM;
+	detect_memory_layout(chunk_array);
+	buf = kzalloc(MEMORY_CHUNKS * CHUNK_INFO_SIZE, GFP_KERNEL);
+	if (!buf) {
+		kfree(chunk_array);
+		return -ENOMEM;
+	}
+	for (i = 0; i < MEMORY_CHUNKS; i++) {
+		sprintf(buf + (i * CHUNK_INFO_SIZE), "%016llx %016llx ",
+			(unsigned long long) chunk_array[i].addr,
+			(unsigned long long) chunk_array[i].size);
+		if (chunk_array[i].size == 0)
+			break;
+	}
+	kfree(chunk_array);
+	filp->private_data = buf;
+	return 0;
+}
+
+static int zcore_memmap_release(struct inode *inode, struct file *filp)
+{
+	kfree(filp->private_data);
+	return 0;
+}
+
+static const struct file_operations zcore_memmap_fops = {
+	.owner		= THIS_MODULE,
+	.read		= zcore_memmap_read,
+	.open		= zcore_memmap_open,
+	.release	= zcore_memmap_release,
+};
+
+static ssize_t zcore_reipl_write(struct file *filp, const char __user *buf,
+				 size_t count, loff_t *ppos)
+{
+	if (ipl_block) {
+		diag308(DIAG308_SET, ipl_block);
+		diag308(DIAG308_IPL, NULL);
+	}
+	return count;
+}
+
+static int zcore_reipl_open(struct inode *inode, struct file *filp)
+{
+	return 0;
+}
+
+static int zcore_reipl_release(struct inode *inode, struct file *filp)
+{
+	return 0;
+}
+
+static const struct file_operations zcore_reipl_fops = {
+	.owner		= THIS_MODULE,
+	.write		= zcore_reipl_write,
+	.open		= zcore_reipl_open,
+	.release	= zcore_reipl_release,
+};
+
+
+static void __init set_s390_lc_mask(union save_area *map)
+{
+	memset(&map->s390.ext_save, 0xff, sizeof(map->s390.ext_save));
+	memset(&map->s390.timer, 0xff, sizeof(map->s390.timer));
+	memset(&map->s390.clk_cmp, 0xff, sizeof(map->s390.clk_cmp));
+	memset(&map->s390.psw, 0xff, sizeof(map->s390.psw));
+	memset(&map->s390.pref_reg, 0xff, sizeof(map->s390.pref_reg));
+	memset(&map->s390.acc_regs, 0xff, sizeof(map->s390.acc_regs));
+	memset(&map->s390.fp_regs, 0xff, sizeof(map->s390.fp_regs));
+	memset(&map->s390.gp_regs, 0xff, sizeof(map->s390.gp_regs));
+	memset(&map->s390.ctrl_regs, 0xff, sizeof(map->s390.ctrl_regs));
+}
+
+static void __init set_s390x_lc_mask(union save_area *map)
+{
+	memset(&map->s390x.fp_regs, 0xff, sizeof(map->s390x.fp_regs));
+	memset(&map->s390x.gp_regs, 0xff, sizeof(map->s390x.gp_regs));
+	memset(&map->s390x.psw, 0xff, sizeof(map->s390x.psw));
+	memset(&map->s390x.pref_reg, 0xff, sizeof(map->s390x.pref_reg));
+	memset(&map->s390x.fp_ctrl_reg, 0xff, sizeof(map->s390x.fp_ctrl_reg));
+	memset(&map->s390x.tod_reg, 0xff, sizeof(map->s390x.tod_reg));
+	memset(&map->s390x.timer, 0xff, sizeof(map->s390x.timer));
+	memset(&map->s390x.clk_cmp, 0xff, sizeof(map->s390x.clk_cmp));
+	memset(&map->s390x.acc_regs, 0xff, sizeof(map->s390x.acc_regs));
+	memset(&map->s390x.ctrl_regs, 0xff, sizeof(map->s390x.ctrl_regs));
+}
+
+/*
+ * Initialize dump globals for a given architecture
+ */
+static int __init sys_info_init(enum arch_id arch)
+{
+	int rc;
+
+	switch (arch) {
+	case ARCH_S390X:
+		pr_alert("DETECTED 'S390X (64 bit) OS'\n");
+		sys_info.sa_base = SAVE_AREA_BASE_S390X;
+		sys_info.sa_size = sizeof(struct save_area_s390x);
+		set_s390x_lc_mask(&sys_info.lc_mask);
+		break;
+	case ARCH_S390:
+		pr_alert("DETECTED 'S390 (32 bit) OS'\n");
+		sys_info.sa_base = SAVE_AREA_BASE_S390;
+		sys_info.sa_size = sizeof(struct save_area_s390);
+		set_s390_lc_mask(&sys_info.lc_mask);
+		break;
+	default:
+		pr_alert("0x%x is an unknown architecture.\n",arch);
+		return -EINVAL;
+	}
+	sys_info.arch = arch;
+	rc = init_cpu_info(arch);
+	if (rc)
+		return rc;
+	sys_info.mem_size = real_memory_size;
+
+	return 0;
+}
+
+static int __init check_sdias(void)
+{
+	int rc, act_hsa_size;
+
+	rc = sclp_sdias_blk_count();
+	if (rc < 0) {
+		TRACE("Could not determine HSA size\n");
+		return rc;
+	}
+	act_hsa_size = (rc - 1) * PAGE_SIZE;
+	if (act_hsa_size < ZFCPDUMP_HSA_SIZE) {
+		TRACE("HSA size too small: %i\n", act_hsa_size);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int __init get_mem_size(unsigned long *mem)
+{
+	int i;
+	struct mem_chunk *chunk_array;
+
+	chunk_array = kzalloc(MEMORY_CHUNKS * sizeof(struct mem_chunk),
+			      GFP_KERNEL);
+	if (!chunk_array)
+		return -ENOMEM;
+	detect_memory_layout(chunk_array);
+	for (i = 0; i < MEMORY_CHUNKS; i++) {
+		if (chunk_array[i].size == 0)
+			break;
+		*mem += chunk_array[i].size;
+	}
+	kfree(chunk_array);
+	return 0;
+}
+
+static int __init zcore_header_init(int arch, struct zcore_header *hdr)
+{
+	int rc;
+	unsigned long memory = 0;
+
+	if (arch == ARCH_S390X)
+		hdr->arch_id = DUMP_ARCH_S390X;
+	else
+		hdr->arch_id = DUMP_ARCH_S390;
+	rc = get_mem_size(&memory);
+	if (rc)
+		return rc;
+	hdr->mem_size = memory;
+	hdr->rmem_size = memory;
+	hdr->mem_end = sys_info.mem_size;
+	hdr->num_pages = memory / PAGE_SIZE;
+	hdr->tod = get_clock();
+	get_cpu_id(&hdr->cpu_id);
+	return 0;
+}
+
+/*
+ * Provide IPL parameter information block from either HSA or memory
+ * for future reipl
+ */
+static int __init zcore_reipl_init(void)
+{
+	struct ipib_info ipib_info;
+	int rc;
+
+	rc = memcpy_hsa_kernel(&ipib_info, __LC_DUMP_REIPL, sizeof(ipib_info));
+	if (rc)
+		return rc;
+	if (ipib_info.ipib == 0)
+		return 0;
+	ipl_block = (void *) __get_free_page(GFP_KERNEL);
+	if (!ipl_block)
+		return -ENOMEM;
+	if (ipib_info.ipib < ZFCPDUMP_HSA_SIZE)
+		rc = memcpy_hsa_kernel(ipl_block, ipib_info.ipib, PAGE_SIZE);
+	else
+		rc = memcpy_real(ipl_block, ipib_info.ipib, PAGE_SIZE);
+	if (rc) {
+		free_page((unsigned long) ipl_block);
+		return rc;
+	}
+	if (csum_partial(ipl_block, ipl_block->hdr.len, 0) !=
+	    ipib_info.checksum) {
+		TRACE("Checksum does not match\n");
+		free_page((unsigned long) ipl_block);
+		ipl_block = NULL;
+	}
+	return 0;
+}
+
+static int __init zcore_init(void)
+{
+	unsigned char arch;
+	int rc;
+
+	if (ipl_info.type != IPL_TYPE_FCP_DUMP)
+		return -ENODATA;
+
+	zcore_dbf = debug_register("zcore", 4, 1, 4 * sizeof(long));
+	debug_register_view(zcore_dbf, &debug_sprintf_view);
+	debug_set_level(zcore_dbf, 6);
+
+	TRACE("devno:  %x\n", ipl_info.data.fcp.dev_id.devno);
+	TRACE("wwpn:   %llx\n", (unsigned long long) ipl_info.data.fcp.wwpn);
+	TRACE("lun:    %llx\n", (unsigned long long) ipl_info.data.fcp.lun);
+
+	rc = sclp_sdias_init();
+	if (rc)
+		goto fail;
+
+	rc = check_sdias();
+	if (rc)
+		goto fail;
+
+	rc = memcpy_hsa_kernel(&arch, __LC_AR_MODE_ID, 1);
+	if (rc)
+		goto fail;
+
+#ifndef __s390x__
+	if (arch == ARCH_S390X) {
+		pr_alert("The 32-bit dump tool cannot be used for a "
+			 "64-bit system\n");
+		rc = -EINVAL;
+		goto fail;
+	}
+#endif
+
+	rc = sys_info_init(arch);
+	if (rc)
+		goto fail;
+
+	rc = zcore_header_init(arch, &zcore_header);
+	if (rc)
+		goto fail;
+
+	rc = zcore_reipl_init();
+	if (rc)
+		goto fail;
+
+	zcore_dir = debugfs_create_dir("zcore" , NULL);
+	if (!zcore_dir) {
+		rc = -ENOMEM;
+		goto fail;
+	}
+	zcore_file = debugfs_create_file("mem", S_IRUSR, zcore_dir, NULL,
+					 &zcore_fops);
+	if (!zcore_file) {
+		rc = -ENOMEM;
+		goto fail_dir;
+	}
+	zcore_memmap_file = debugfs_create_file("memmap", S_IRUSR, zcore_dir,
+						NULL, &zcore_memmap_fops);
+	if (!zcore_memmap_file) {
+		rc = -ENOMEM;
+		goto fail_file;
+	}
+	zcore_reipl_file = debugfs_create_file("reipl", S_IRUSR, zcore_dir,
+						NULL, &zcore_reipl_fops);
+	if (!zcore_reipl_file) {
+		rc = -ENOMEM;
+		goto fail_memmap_file;
+	}
+	hsa_available = 1;
+	return 0;
+
+fail_memmap_file:
+	debugfs_remove(zcore_memmap_file);
+fail_file:
+	debugfs_remove(zcore_file);
+fail_dir:
+	debugfs_remove(zcore_dir);
+fail:
+	diag308(DIAG308_REL_HSA, NULL);
+	return rc;
+}
+
+static void __exit zcore_exit(void)
+{
+	debug_unregister(zcore_dbf);
+	sclp_sdias_exit();
+	free_page((unsigned long) ipl_block);
+	debugfs_remove(zcore_reipl_file);
+	debugfs_remove(zcore_memmap_file);
+	debugfs_remove(zcore_file);
+	debugfs_remove(zcore_dir);
+	diag308(DIAG308_REL_HSA, NULL);
+}
+
+MODULE_AUTHOR("Copyright IBM Corp. 2003,2008");
+MODULE_DESCRIPTION("zcore module for zfcpdump support");
+MODULE_LICENSE("GPL");
+
+subsys_initcall(zcore_init);
+module_exit(zcore_exit);

kernel/drivers/s390/cio/Makefile

@@ -0,0 +1,14 @@
+#
+# Makefile for the S/390 common i/o drivers
+#
+
+obj-y += airq.o blacklist.o chsc.o cio.o css.o chp.o idset.o isc.o \
+	fcx.o itcw.o crw.o
+ccw_device-objs += device.o device_fsm.o device_ops.o
+ccw_device-objs += device_id.o device_pgid.o device_status.o
+obj-y += ccw_device.o cmf.o
+obj-$(CONFIG_CHSC_SCH) += chsc_sch.o
+obj-$(CONFIG_CCWGROUP) += ccwgroup.o
+
+qdio-objs := qdio_main.o qdio_thinint.o qdio_debug.o qdio_perf.o qdio_setup.o
+obj-$(CONFIG_QDIO) += qdio.o

kernel/drivers/s390/cio/airq.c

@@ -0,0 +1,151 @@
+/*
+ *  drivers/s390/cio/airq.c
+ *    Support for adapter interruptions
+ *
+ *    Copyright IBM Corp. 1999,2007
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>
+ *		 Arnd Bergmann <arndb@de.ibm.com>
+ *		 Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/rcupdate.h>
+
+#include <asm/airq.h>
+#include <asm/isc.h>
+
+#include "cio.h"
+#include "cio_debug.h"
+
+#define NR_AIRQS		32
+#define NR_AIRQS_PER_WORD	sizeof(unsigned long)
+#define NR_AIRQ_WORDS		(NR_AIRQS / NR_AIRQS_PER_WORD)
+
+union indicator_t {
+	unsigned long word[NR_AIRQ_WORDS];
+	unsigned char byte[NR_AIRQS];
+} __attribute__((packed));
+
+struct airq_t {
+	adapter_int_handler_t handler;
+	void *drv_data;
+};
+
+static union indicator_t indicators[MAX_ISC+1];
+static struct airq_t *airqs[MAX_ISC+1][NR_AIRQS];
+
+static int register_airq(struct airq_t *airq, u8 isc)
+{
+	int i;
+
+	for (i = 0; i < NR_AIRQS; i++)
+		if (!cmpxchg(&airqs[isc][i], NULL, airq))
+			return i;
+	return -ENOMEM;
+}
+
+/**
+ * s390_register_adapter_interrupt() - register adapter interrupt handler
+ * @handler: adapter handler to be registered
+ * @drv_data: driver data passed with each call to the handler
+ * @isc: isc for which the handler should be called
+ *
+ * Returns:
+ *  Pointer to the indicator to be used on success
+ *  ERR_PTR() if registration failed
+ */
+void *s390_register_adapter_interrupt(adapter_int_handler_t handler,
+				      void *drv_data, u8 isc)
+{
+	struct airq_t *airq;
+	char dbf_txt[16];
+	int ret;
+
+	if (isc > MAX_ISC)
+		return ERR_PTR(-EINVAL);
+	airq = kmalloc(sizeof(struct airq_t), GFP_KERNEL);
+	if (!airq) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	airq->handler = handler;
+	airq->drv_data = drv_data;
+
+	ret = register_airq(airq, isc);
+out:
+	snprintf(dbf_txt, sizeof(dbf_txt), "rairq:%d", ret);
+	CIO_TRACE_EVENT(4, dbf_txt);
+	if (ret < 0) {
+		kfree(airq);
+		return ERR_PTR(ret);
+	} else
+		return &indicators[isc].byte[ret];
+}
+EXPORT_SYMBOL(s390_register_adapter_interrupt);
+
+/**
+ * s390_unregister_adapter_interrupt - unregister adapter interrupt handler
+ * @ind: indicator for which the handler is to be unregistered
+ * @isc: interruption subclass
+ */
+void s390_unregister_adapter_interrupt(void *ind, u8 isc)
+{
+	struct airq_t *airq;
+	char dbf_txt[16];
+	int i;
+
+	i = (int) ((addr_t) ind) - ((addr_t) &indicators[isc].byte[0]);
+	snprintf(dbf_txt, sizeof(dbf_txt), "urairq:%d", i);
+	CIO_TRACE_EVENT(4, dbf_txt);
+	indicators[isc].byte[i] = 0;
+	airq = xchg(&airqs[isc][i], NULL);
+	/*
+	 * Allow interrupts to complete. This will ensure that the airq handle
+	 * is no longer referenced by any interrupt handler.
+	 */
+	synchronize_sched();
+	kfree(airq);
+}
+EXPORT_SYMBOL(s390_unregister_adapter_interrupt);
+
+#define INDICATOR_MASK	(0xffUL << ((NR_AIRQS_PER_WORD - 1) * 8))
+
+void do_adapter_IO(u8 isc)
+{
+	int w;
+	int i;
+	unsigned long word;
+	struct airq_t *airq;
+
+	/*
+	 * Access indicator array in word-sized chunks to minimize storage
+	 * fetch operations.
+	 */
+	for (w = 0; w < NR_AIRQ_WORDS; w++) {
+		word = indicators[isc].word[w];
+		i = w * NR_AIRQS_PER_WORD;
+		/*
+		 * Check bytes within word for active indicators.
+		 */
+		while (word) {
+			if (word & INDICATOR_MASK) {
+				airq = airqs[isc][i];
+				/* Make sure gcc reads from airqs only once. */
+				barrier();
+				if (likely(airq))
+					airq->handler(&indicators[isc].byte[i],
+						      airq->drv_data);
+				else
+					/*
+					 * Reset ill-behaved indicator.
+					 */
+					indicators[isc].byte[i] = 0;
+			}
+			word <<= 8;
+			i++;
+		}
+	}
+}

kernel/drivers/s390/cio/blacklist.c

@@ -0,0 +1,403 @@
+/*
+ *  drivers/s390/cio/blacklist.c
+ *   S/390 common I/O routines -- blacklisting of specific devices
+ *
+ *    Copyright (C) 1999-2002 IBM Deutschland Entwicklung GmbH,
+ *			      IBM Corporation
+ *    Author(s): Ingo Adlung (adlung@de.ibm.com)
+ *		 Cornelia Huck (cornelia.huck@de.ibm.com)
+ *		 Arnd Bergmann (arndb@de.ibm.com)
+ */
+
+#define KMSG_COMPONENT "cio"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/init.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/ctype.h>
+#include <linux/device.h>
+
+#include <asm/cio.h>
+#include <asm/uaccess.h>
+
+#include "blacklist.h"
+#include "cio.h"
+#include "cio_debug.h"
+#include "css.h"
+#include "device.h"
+
+/*
+ * "Blacklisting" of certain devices:
+ * Device numbers given in the commandline as cio_ignore=... won't be known
+ * to Linux.
+ *
+ * These can be single devices or ranges of devices
+ */
+
+/* 65536 bits for each set to indicate if a devno is blacklisted or not */
+#define __BL_DEV_WORDS ((__MAX_SUBCHANNEL + (8*sizeof(long) - 1)) / \
+			 (8*sizeof(long)))
+static unsigned long bl_dev[__MAX_SSID + 1][__BL_DEV_WORDS];
+typedef enum {add, free} range_action;
+
+/*
+ * Function: blacklist_range
+ * (Un-)blacklist the devices from-to
+ */
+static int blacklist_range(range_action action, unsigned int from_ssid,
+			   unsigned int to_ssid, unsigned int from,
+			   unsigned int to, int msgtrigger)
+{
+	if ((from_ssid > to_ssid) || ((from_ssid == to_ssid) && (from > to))) {
+		if (msgtrigger)
+			pr_warning("0.%x.%04x to 0.%x.%04x is not a valid "
+				   "range for cio_ignore\n", from_ssid, from,
+				   to_ssid, to);
+
+		return 1;
+	}
+
+	while ((from_ssid < to_ssid) || ((from_ssid == to_ssid) &&
+	       (from <= to))) {
+		if (action == add)
+			set_bit(from, bl_dev[from_ssid]);
+		else
+			clear_bit(from, bl_dev[from_ssid]);
+		from++;
+		if (from > __MAX_SUBCHANNEL) {
+			from_ssid++;
+			from = 0;
+		}
+	}
+
+	return 0;
+}
+
+static int pure_hex(char **cp, unsigned int *val, int min_digit,
+		    int max_digit, int max_val)
+{
+	int diff;
+	unsigned int value;
+
+	diff = 0;
+	*val = 0;
+
+	while (isxdigit(**cp) && (diff <= max_digit)) {
+
+		if (isdigit(**cp))
+			value = **cp - '0';
+		else
+			value = tolower(**cp) - 'a' + 10;
+		*val = *val * 16 + value;
+		(*cp)++;
+		diff++;
+	}
+
+	if ((diff < min_digit) || (diff > max_digit) || (*val > max_val))
+		return 1;
+
+	return 0;
+}
+
+static int parse_busid(char *str, unsigned int *cssid, unsigned int *ssid,
+		       unsigned int *devno, int msgtrigger)
+{
+	char *str_work;
+	int val, rc, ret;
+
+	rc = 1;
+
+	if (*str == '\0')
+		goto out;
+
+	/* old style */
+	str_work = str;
+	val = simple_strtoul(str, &str_work, 16);
+
+	if (*str_work == '\0') {
+		if (val <= __MAX_SUBCHANNEL) {
+			*devno = val;
+			*ssid = 0;
+			*cssid = 0;
+			rc = 0;
+		}
+		goto out;
+	}
+
+	/* new style */
+	str_work = str;
+	ret = pure_hex(&str_work, cssid, 1, 2, __MAX_CSSID);
+	if (ret || (str_work[0] != '.'))
+		goto out;
+	str_work++;
+	ret = pure_hex(&str_work, ssid, 1, 1, __MAX_SSID);
+	if (ret || (str_work[0] != '.'))
+		goto out;
+	str_work++;
+	ret = pure_hex(&str_work, devno, 4, 4, __MAX_SUBCHANNEL);
+	if (ret || (str_work[0] != '\0'))
+		goto out;
+
+	rc = 0;
+out:
+	if (rc && msgtrigger)
+		pr_warning("%s is not a valid device for the cio_ignore "
+			   "kernel parameter\n", str);
+
+	return rc;
+}
+
+static int blacklist_parse_parameters(char *str, range_action action,
+				      int msgtrigger)
+{
+	unsigned int from_cssid, to_cssid, from_ssid, to_ssid, from, to;
+	int rc, totalrc;
+	char *parm;
+	range_action ra;
+
+	totalrc = 0;
+
+	while ((parm = strsep(&str, ","))) {
+		rc = 0;
+		ra = action;
+		if (*parm == '!') {
+			if (ra == add)
+				ra = free;
+			else
+				ra = add;
+			parm++;
+		}
+		if (strcmp(parm, "all") == 0) {
+			from_cssid = 0;
+			from_ssid = 0;
+			from = 0;
+			to_cssid = __MAX_CSSID;
+			to_ssid = __MAX_SSID;
+			to = __MAX_SUBCHANNEL;
+		} else {
+			rc = parse_busid(strsep(&parm, "-"), &from_cssid,
+					 &from_ssid, &from, msgtrigger);
+			if (!rc) {
+				if (parm != NULL)
+					rc = parse_busid(parm, &to_cssid,
+							 &to_ssid, &to,
+							 msgtrigger);
+				else {
+					to_cssid = from_cssid;
+					to_ssid = from_ssid;
+					to = from;
+				}
+			}
+		}
+		if (!rc) {
+			rc = blacklist_range(ra, from_ssid, to_ssid, from, to,
+					     msgtrigger);
+			if (rc)
+				totalrc = -EINVAL;
+		} else
+			totalrc = -EINVAL;
+	}
+
+	return totalrc;
+}
+
+static int __init
+blacklist_setup (char *str)
+{
+	CIO_MSG_EVENT(6, "Reading blacklist parameters\n");
+	if (blacklist_parse_parameters(str, add, 1))
+		return 0;
+	return 1;
+}
+
+__setup ("cio_ignore=", blacklist_setup);
+
+/* Checking if devices are blacklisted */
+
+/*
+ * Function: is_blacklisted
+ * Returns 1 if the given devicenumber can be found in the blacklist,
+ * otherwise 0.
+ * Used by validate_subchannel()
+ */
+int
+is_blacklisted (int ssid, int devno)
+{
+	return test_bit (devno, bl_dev[ssid]);
+}
+
+#ifdef CONFIG_PROC_FS
+/*
+ * Function: blacklist_parse_proc_parameters
+ * parse the stuff which is piped to /proc/cio_ignore
+ */
+static int blacklist_parse_proc_parameters(char *buf)
+{
+	int rc;
+	char *parm;
+
+	parm = strsep(&buf, " ");
+
+	if (strcmp("free", parm) == 0)
+		rc = blacklist_parse_parameters(buf, free, 0);
+	else if (strcmp("add", parm) == 0)
+		rc = blacklist_parse_parameters(buf, add, 0);
+	else if (strcmp("purge", parm) == 0)
+		return ccw_purge_blacklisted();
+	else
+		return -EINVAL;
+
+	css_schedule_reprobe();
+
+	return rc;
+}
+
+/* Iterator struct for all devices. */
+struct ccwdev_iter {
+	int devno;
+	int ssid;
+	int in_range;
+};
+
+static void *
+cio_ignore_proc_seq_start(struct seq_file *s, loff_t *offset)
+{
+	struct ccwdev_iter *iter = s->private;
+
+	if (*offset >= (__MAX_SUBCHANNEL + 1) * (__MAX_SSID + 1))
+		return NULL;
+	memset(iter, 0, sizeof(*iter));
+	iter->ssid = *offset / (__MAX_SUBCHANNEL + 1);
+	iter->devno = *offset % (__MAX_SUBCHANNEL + 1);
+	return iter;
+}
+
+static void
+cio_ignore_proc_seq_stop(struct seq_file *s, void *it)
+{
+}
+
+static void *
+cio_ignore_proc_seq_next(struct seq_file *s, void *it, loff_t *offset)
+{
+	struct ccwdev_iter *iter;
+
+	if (*offset >= (__MAX_SUBCHANNEL + 1) * (__MAX_SSID + 1))
+		return NULL;
+	iter = it;
+	if (iter->devno == __MAX_SUBCHANNEL) {
+		iter->devno = 0;
+		iter->ssid++;
+		if (iter->ssid > __MAX_SSID)
+			return NULL;
+	} else
+		iter->devno++;
+	(*offset)++;
+	return iter;
+}
+
+static int
+cio_ignore_proc_seq_show(struct seq_file *s, void *it)
+{
+	struct ccwdev_iter *iter;
+
+	iter = it;
+	if (!is_blacklisted(iter->ssid, iter->devno))
+		/* Not blacklisted, nothing to output. */
+		return 0;
+	if (!iter->in_range) {
+		/* First device in range. */
+		if ((iter->devno == __MAX_SUBCHANNEL) ||
+		    !is_blacklisted(iter->ssid, iter->devno + 1))
+			/* Singular device. */
+			return seq_printf(s, "0.%x.%04x\n",
+					  iter->ssid, iter->devno);
+		iter->in_range = 1;
+		return seq_printf(s, "0.%x.%04x-", iter->ssid, iter->devno);
+	}
+	if ((iter->devno == __MAX_SUBCHANNEL) ||
+	    !is_blacklisted(iter->ssid, iter->devno + 1)) {
+		/* Last device in range. */
+		iter->in_range = 0;
+		return seq_printf(s, "0.%x.%04x\n", iter->ssid, iter->devno);
+	}
+	return 0;
+}
+
+static ssize_t
+cio_ignore_write(struct file *file, const char __user *user_buf,
+		 size_t user_len, loff_t *offset)
+{
+	char *buf;
+	ssize_t rc, ret, i;
+
+	if (*offset)
+		return -EINVAL;
+	if (user_len > 65536)
+		user_len = 65536;
+	buf = vmalloc (user_len + 1); /* maybe better use the stack? */
+	if (buf == NULL)
+		return -ENOMEM;
+	memset(buf, 0, user_len + 1);
+
+	if (strncpy_from_user (buf, user_buf, user_len) < 0) {
+		rc = -EFAULT;
+		goto out_free;
+	}
+
+	i = user_len - 1;
+	while ((i >= 0) && (isspace(buf[i]) || (buf[i] == 0))) {
+		buf[i] = '\0';
+		i--;
+	}
+	ret = blacklist_parse_proc_parameters(buf);
+	if (ret)
+		rc = ret;
+	else
+		rc = user_len;
+
+out_free:
+	vfree (buf);
+	return rc;
+}
+
+static const struct seq_operations cio_ignore_proc_seq_ops = {
+	.start = cio_ignore_proc_seq_start,
+	.stop  = cio_ignore_proc_seq_stop,
+	.next  = cio_ignore_proc_seq_next,
+	.show  = cio_ignore_proc_seq_show,
+};
+
+static int
+cio_ignore_proc_open(struct inode *inode, struct file *file)
+{
+	return seq_open_private(file, &cio_ignore_proc_seq_ops,
+				sizeof(struct ccwdev_iter));
+}
+
+static const struct file_operations cio_ignore_proc_fops = {
+	.open    = cio_ignore_proc_open,
+	.read    = seq_read,
+	.llseek  = seq_lseek,
+	.release = seq_release_private,
+	.write   = cio_ignore_write,
+};
+
+static int
+cio_ignore_proc_init (void)
+{
+	struct proc_dir_entry *entry;
+
+	entry = proc_create("cio_ignore", S_IFREG | S_IRUGO | S_IWUSR, NULL,
+			    &cio_ignore_proc_fops);
+	if (!entry)
+		return -ENOENT;
+	return 0;
+}
+
+__initcall (cio_ignore_proc_init);
+
+#endif /* CONFIG_PROC_FS */

kernel/drivers/s390/cio/blacklist.h

@@ -0,0 +1,6 @@
+#ifndef S390_BLACKLIST_H
+#define S390_BLACKLIST_H
+
+extern int is_blacklisted (int ssid, int devno);
+
+#endif

kernel/drivers/s390/cio/ccwgroup.c

@@ -0,0 +1,703 @@
+/*
+ *  bus driver for ccwgroup
+ *
+ *  Copyright IBM Corp. 2002, 2009
+ *
+ *  Author(s): Arnd Bergmann (arndb@de.ibm.com)
+ *	       Cornelia Huck (cornelia.huck@de.ibm.com)
+ */
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/list.h>
+#include <linux/device.h>
+#include <linux/init.h>
+#include <linux/ctype.h>
+#include <linux/dcache.h>
+
+#include <asm/ccwdev.h>
+#include <asm/ccwgroup.h>
+
+#define CCW_BUS_ID_SIZE		20
+
+/* In Linux 2.4, we had a channel device layer called "chandev"
+ * that did all sorts of obscure stuff for networking devices.
+ * This is another driver that serves as a replacement for just
+ * one of its functions, namely the translation of single subchannels
+ * to devices that use multiple subchannels.
+ */
+
+/* a device matches a driver if all its slave devices match the same
+ * entry of the driver */
+static int
+ccwgroup_bus_match (struct device * dev, struct device_driver * drv)
+{
+	struct ccwgroup_device *gdev;
+	struct ccwgroup_driver *gdrv;
+
+	gdev = to_ccwgroupdev(dev);
+	gdrv = to_ccwgroupdrv(drv);
+
+	if (gdev->creator_id == gdrv->driver_id)
+		return 1;
+
+	return 0;
+}
+static int
+ccwgroup_uevent (struct device *dev, struct kobj_uevent_env *env)
+{
+	/* TODO */
+	return 0;
+}
+
+static struct bus_type ccwgroup_bus_type;
+
+static void
+__ccwgroup_remove_symlinks(struct ccwgroup_device *gdev)
+{
+	int i;
+	char str[8];
+
+	for (i = 0; i < gdev->count; i++) {
+		sprintf(str, "cdev%d", i);
+		sysfs_remove_link(&gdev->dev.kobj, str);
+		sysfs_remove_link(&gdev->cdev[i]->dev.kobj, "group_device");
+	}
+	
+}
+
+/*
+ * Provide an 'ungroup' attribute so the user can remove group devices no
+ * longer needed or accidentially created. Saves memory :)
+ */
+static void ccwgroup_ungroup_callback(struct device *dev)
+{
+	struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
+
+	mutex_lock(&gdev->reg_mutex);
+	if (device_is_registered(&gdev->dev)) {
+		__ccwgroup_remove_symlinks(gdev);
+		device_unregister(dev);
+	}
+	mutex_unlock(&gdev->reg_mutex);
+}
+
+static ssize_t
+ccwgroup_ungroup_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct ccwgroup_device *gdev;
+	int rc;
+
+	gdev = to_ccwgroupdev(dev);
+
+	/* Prevent concurrent online/offline processing and ungrouping. */
+	if (atomic_cmpxchg(&gdev->onoff, 0, 1) != 0)
+		return -EAGAIN;
+	if (gdev->state != CCWGROUP_OFFLINE) {
+		rc = -EINVAL;
+		goto out;
+	}
+	/* Note that we cannot unregister the device from one of its
+	 * attribute methods, so we have to use this roundabout approach.
+	 */
+	rc = device_schedule_callback(dev, ccwgroup_ungroup_callback);
+out:
+	if (rc) {
+		if (rc != -EAGAIN)
+			/* Release onoff "lock" when ungrouping failed. */
+			atomic_set(&gdev->onoff, 0);
+		return rc;
+	}
+	return count;
+}
+
+static DEVICE_ATTR(ungroup, 0200, NULL, ccwgroup_ungroup_store);
+
+static void
+ccwgroup_release (struct device *dev)
+{
+	struct ccwgroup_device *gdev;
+	int i;
+
+	gdev = to_ccwgroupdev(dev);
+
+	for (i = 0; i < gdev->count; i++) {
+		if (gdev->cdev[i]) {
+			if (dev_get_drvdata(&gdev->cdev[i]->dev) == gdev)
+				dev_set_drvdata(&gdev->cdev[i]->dev, NULL);
+			put_device(&gdev->cdev[i]->dev);
+		}
+	}
+	kfree(gdev);
+}
+
+static int
+__ccwgroup_create_symlinks(struct ccwgroup_device *gdev)
+{
+	char str[8];
+	int i, rc;
+
+	for (i = 0; i < gdev->count; i++) {
+		rc = sysfs_create_link(&gdev->cdev[i]->dev.kobj, &gdev->dev.kobj,
+				       "group_device");
+		if (rc) {
+			for (--i; i >= 0; i--)
+				sysfs_remove_link(&gdev->cdev[i]->dev.kobj,
+						  "group_device");
+			return rc;
+		}
+	}
+	for (i = 0; i < gdev->count; i++) {
+		sprintf(str, "cdev%d", i);
+		rc = sysfs_create_link(&gdev->dev.kobj, &gdev->cdev[i]->dev.kobj,
+				       str);
+		if (rc) {
+			for (--i; i >= 0; i--) {
+				sprintf(str, "cdev%d", i);
+				sysfs_remove_link(&gdev->dev.kobj, str);
+			}
+			for (i = 0; i < gdev->count; i++)
+				sysfs_remove_link(&gdev->cdev[i]->dev.kobj,
+						  "group_device");
+			return rc;
+		}
+	}
+	return 0;
+}
+
+static int __get_next_bus_id(const char **buf, char *bus_id)
+{
+	int rc, len;
+	char *start, *end;
+
+	start = (char *)*buf;
+	end = strchr(start, ',');
+	if (!end) {
+		/* Last entry. Strip trailing newline, if applicable. */
+		end = strchr(start, '\n');
+		if (end)
+			*end = '\0';
+		len = strlen(start) + 1;
+	} else {
+		len = end - start + 1;
+		end++;
+	}
+	if (len < CCW_BUS_ID_SIZE) {
+		strlcpy(bus_id, start, len);
+		rc = 0;
+	} else
+		rc = -EINVAL;
+	*buf = end;
+	return rc;
+}
+
+static int __is_valid_bus_id(char bus_id[CCW_BUS_ID_SIZE])
+{
+	int cssid, ssid, devno;
+
+	/* Must be of form %x.%x.%04x */
+	if (sscanf(bus_id, "%x.%1x.%04x", &cssid, &ssid, &devno) != 3)
+		return 0;
+	return 1;
+}
+
+/**
+ * ccwgroup_create_from_string() - create and register a ccw group device
+ * @root: parent device for the new device
+ * @creator_id: identifier of creating driver
+ * @cdrv: ccw driver of slave devices
+ * @num_devices: number of slave devices
+ * @buf: buffer containing comma separated bus ids of slave devices
+ *
+ * Create and register a new ccw group device as a child of @root. Slave
+ * devices are obtained from the list of bus ids given in @buf and must all
+ * belong to @cdrv.
+ * Returns:
+ *  %0 on success and an error code on failure.
+ * Context:
+ *  non-atomic
+ */
+int ccwgroup_create_from_string(struct device *root, unsigned int creator_id,
+				struct ccw_driver *cdrv, int num_devices,
+				const char *buf)
+{
+	struct ccwgroup_device *gdev;
+	int rc, i;
+	char tmp_bus_id[CCW_BUS_ID_SIZE];
+	const char *curr_buf;
+
+	gdev = kzalloc(sizeof(*gdev) + num_devices * sizeof(gdev->cdev[0]),
+		       GFP_KERNEL);
+	if (!gdev)
+		return -ENOMEM;
+
+	atomic_set(&gdev->onoff, 0);
+	mutex_init(&gdev->reg_mutex);
+	mutex_lock(&gdev->reg_mutex);
+	gdev->creator_id = creator_id;
+	gdev->count = num_devices;
+	gdev->dev.bus = &ccwgroup_bus_type;
+	gdev->dev.parent = root;
+	gdev->dev.release = ccwgroup_release;
+	device_initialize(&gdev->dev);
+
+	curr_buf = buf;
+	for (i = 0; i < num_devices && curr_buf; i++) {
+		rc = __get_next_bus_id(&curr_buf, tmp_bus_id);
+		if (rc != 0)
+			goto error;
+		if (!__is_valid_bus_id(tmp_bus_id)) {
+			rc = -EINVAL;
+			goto error;
+		}
+		gdev->cdev[i] = get_ccwdev_by_busid(cdrv, tmp_bus_id);
+		/*
+		 * All devices have to be of the same type in
+		 * order to be grouped.
+		 */
+		if (!gdev->cdev[i]
+		    || gdev->cdev[i]->id.driver_info !=
+		    gdev->cdev[0]->id.driver_info) {
+			rc = -EINVAL;
+			goto error;
+		}
+		/* Don't allow a device to belong to more than one group. */
+		if (dev_get_drvdata(&gdev->cdev[i]->dev)) {
+			rc = -EINVAL;
+			goto error;
+		}
+		dev_set_drvdata(&gdev->cdev[i]->dev, gdev);
+	}
+	/* Check for sufficient number of bus ids. */
+	if (i < num_devices && !curr_buf) {
+		rc = -EINVAL;
+		goto error;
+	}
+	/* Check for trailing stuff. */
+	if (i == num_devices && strlen(curr_buf) > 0) {
+		rc = -EINVAL;
+		goto error;
+	}
+
+	dev_set_name(&gdev->dev, "%s", dev_name(&gdev->cdev[0]->dev));
+
+	rc = device_add(&gdev->dev);
+	if (rc)
+		goto error;
+	get_device(&gdev->dev);
+	rc = device_create_file(&gdev->dev, &dev_attr_ungroup);
+
+	if (rc) {
+		device_unregister(&gdev->dev);
+		goto error;
+	}
+
+	rc = __ccwgroup_create_symlinks(gdev);
+	if (!rc) {
+		mutex_unlock(&gdev->reg_mutex);
+		put_device(&gdev->dev);
+		return 0;
+	}
+	device_remove_file(&gdev->dev, &dev_attr_ungroup);
+	device_unregister(&gdev->dev);
+error:
+	for (i = 0; i < num_devices; i++)
+		if (gdev->cdev[i]) {
+			if (dev_get_drvdata(&gdev->cdev[i]->dev) == gdev)
+				dev_set_drvdata(&gdev->cdev[i]->dev, NULL);
+			put_device(&gdev->cdev[i]->dev);
+			gdev->cdev[i] = NULL;
+		}
+	mutex_unlock(&gdev->reg_mutex);
+	put_device(&gdev->dev);
+	return rc;
+}
+EXPORT_SYMBOL(ccwgroup_create_from_string);
+
+static int ccwgroup_notifier(struct notifier_block *nb, unsigned long action,
+			     void *data);
+
+static struct notifier_block ccwgroup_nb = {
+	.notifier_call = ccwgroup_notifier
+};
+
+static int __init init_ccwgroup(void)
+{
+	int ret;
+
+	ret = bus_register(&ccwgroup_bus_type);
+	if (ret)
+		return ret;
+
+	ret = bus_register_notifier(&ccwgroup_bus_type, &ccwgroup_nb);
+	if (ret)
+		bus_unregister(&ccwgroup_bus_type);
+
+	return ret;
+}
+
+static void __exit cleanup_ccwgroup(void)
+{
+	bus_unregister_notifier(&ccwgroup_bus_type, &ccwgroup_nb);
+	bus_unregister(&ccwgroup_bus_type);
+}
+
+module_init(init_ccwgroup);
+module_exit(cleanup_ccwgroup);
+
+/************************** driver stuff ******************************/
+
+static int
+ccwgroup_set_online(struct ccwgroup_device *gdev)
+{
+	struct ccwgroup_driver *gdrv;
+	int ret;
+
+	if (atomic_cmpxchg(&gdev->onoff, 0, 1) != 0)
+		return -EAGAIN;
+	if (gdev->state == CCWGROUP_ONLINE) {
+		ret = 0;
+		goto out;
+	}
+	if (!gdev->dev.driver) {
+		ret = -EINVAL;
+		goto out;
+	}
+	gdrv = to_ccwgroupdrv (gdev->dev.driver);
+	if ((ret = gdrv->set_online ? gdrv->set_online(gdev) : 0))
+		goto out;
+
+	gdev->state = CCWGROUP_ONLINE;
+ out:
+	atomic_set(&gdev->onoff, 0);
+	return ret;
+}
+
+static int
+ccwgroup_set_offline(struct ccwgroup_device *gdev)
+{
+	struct ccwgroup_driver *gdrv;
+	int ret;
+
+	if (atomic_cmpxchg(&gdev->onoff, 0, 1) != 0)
+		return -EAGAIN;
+	if (gdev->state == CCWGROUP_OFFLINE) {
+		ret = 0;
+		goto out;
+	}
+	if (!gdev->dev.driver) {
+		ret = -EINVAL;
+		goto out;
+	}
+	gdrv = to_ccwgroupdrv (gdev->dev.driver);
+	if ((ret = gdrv->set_offline ? gdrv->set_offline(gdev) : 0))
+		goto out;
+
+	gdev->state = CCWGROUP_OFFLINE;
+ out:
+	atomic_set(&gdev->onoff, 0);
+	return ret;
+}
+
+static ssize_t
+ccwgroup_online_store (struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct ccwgroup_device *gdev;
+	struct ccwgroup_driver *gdrv;
+	unsigned long value;
+	int ret;
+
+	if (!dev->driver)
+		return -ENODEV;
+
+	gdev = to_ccwgroupdev(dev);
+	gdrv = to_ccwgroupdrv(dev->driver);
+
+	if (!try_module_get(gdrv->owner))
+		return -EINVAL;
+
+	ret = strict_strtoul(buf, 0, &value);
+	if (ret)
+		goto out;
+
+	if (value == 1)
+		ret = ccwgroup_set_online(gdev);
+	else if (value == 0)
+		ret = ccwgroup_set_offline(gdev);
+	else
+		ret = -EINVAL;
+out:
+	module_put(gdrv->owner);
+	return (ret == 0) ? count : ret;
+}
+
+static ssize_t
+ccwgroup_online_show (struct device *dev, struct device_attribute *attr, char *buf)
+{
+	int online;
+
+	online = (to_ccwgroupdev(dev)->state == CCWGROUP_ONLINE);
+
+	return sprintf(buf, online ? "1\n" : "0\n");
+}
+
+static DEVICE_ATTR(online, 0644, ccwgroup_online_show, ccwgroup_online_store);
+
+static int
+ccwgroup_probe (struct device *dev)
+{
+	struct ccwgroup_device *gdev;
+	struct ccwgroup_driver *gdrv;
+
+	int ret;
+
+	gdev = to_ccwgroupdev(dev);
+	gdrv = to_ccwgroupdrv(dev->driver);
+
+	if ((ret = device_create_file(dev, &dev_attr_online)))
+		return ret;
+
+	ret = gdrv->probe ? gdrv->probe(gdev) : -ENODEV;
+	if (ret)
+		device_remove_file(dev, &dev_attr_online);
+
+	return ret;
+}
+
+static int
+ccwgroup_remove (struct device *dev)
+{
+	struct ccwgroup_device *gdev;
+	struct ccwgroup_driver *gdrv;
+
+	device_remove_file(dev, &dev_attr_online);
+	device_remove_file(dev, &dev_attr_ungroup);
+
+	if (!dev->driver)
+		return 0;
+
+	gdev = to_ccwgroupdev(dev);
+	gdrv = to_ccwgroupdrv(dev->driver);
+
+	if (gdrv->remove)
+		gdrv->remove(gdev);
+
+	return 0;
+}
+
+static void ccwgroup_shutdown(struct device *dev)
+{
+	struct ccwgroup_device *gdev;
+	struct ccwgroup_driver *gdrv;
+
+	if (!dev->driver)
+		return;
+
+	gdev = to_ccwgroupdev(dev);
+	gdrv = to_ccwgroupdrv(dev->driver);
+
+	if (gdrv->shutdown)
+		gdrv->shutdown(gdev);
+}
+
+static int ccwgroup_pm_prepare(struct device *dev)
+{
+	struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
+	struct ccwgroup_driver *gdrv = to_ccwgroupdrv(gdev->dev.driver);
+
+	/* Fail while device is being set online/offline. */
+	if (atomic_read(&gdev->onoff))
+		return -EAGAIN;
+
+	if (!gdev->dev.driver || gdev->state != CCWGROUP_ONLINE)
+		return 0;
+
+	return gdrv->prepare ? gdrv->prepare(gdev) : 0;
+}
+
+static void ccwgroup_pm_complete(struct device *dev)
+{
+	struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
+	struct ccwgroup_driver *gdrv = to_ccwgroupdrv(dev->driver);
+
+	if (!gdev->dev.driver || gdev->state != CCWGROUP_ONLINE)
+		return;
+
+	if (gdrv->complete)
+		gdrv->complete(gdev);
+}
+
+static int ccwgroup_pm_freeze(struct device *dev)
+{
+	struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
+	struct ccwgroup_driver *gdrv = to_ccwgroupdrv(gdev->dev.driver);
+
+	if (!gdev->dev.driver || gdev->state != CCWGROUP_ONLINE)
+		return 0;
+
+	return gdrv->freeze ? gdrv->freeze(gdev) : 0;
+}
+
+static int ccwgroup_pm_thaw(struct device *dev)
+{
+	struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
+	struct ccwgroup_driver *gdrv = to_ccwgroupdrv(gdev->dev.driver);
+
+	if (!gdev->dev.driver || gdev->state != CCWGROUP_ONLINE)
+		return 0;
+
+	return gdrv->thaw ? gdrv->thaw(gdev) : 0;
+}
+
+static int ccwgroup_pm_restore(struct device *dev)
+{
+	struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
+	struct ccwgroup_driver *gdrv = to_ccwgroupdrv(gdev->dev.driver);
+
+	if (!gdev->dev.driver || gdev->state != CCWGROUP_ONLINE)
+		return 0;
+
+	return gdrv->restore ? gdrv->restore(gdev) : 0;
+}
+
+static struct dev_pm_ops ccwgroup_pm_ops = {
+	.prepare = ccwgroup_pm_prepare,
+	.complete = ccwgroup_pm_complete,
+	.freeze = ccwgroup_pm_freeze,
+	.thaw = ccwgroup_pm_thaw,
+	.restore = ccwgroup_pm_restore,
+};
+
+static struct bus_type ccwgroup_bus_type = {
+	.name   = "ccwgroup",
+	.match  = ccwgroup_bus_match,
+	.uevent = ccwgroup_uevent,
+	.probe  = ccwgroup_probe,
+	.remove = ccwgroup_remove,
+	.shutdown = ccwgroup_shutdown,
+	.pm = &ccwgroup_pm_ops,
+};
+
+
+static int ccwgroup_notifier(struct notifier_block *nb, unsigned long action,
+			     void *data)
+{
+	struct device *dev = data;
+
+	if (action == BUS_NOTIFY_UNBIND_DRIVER)
+		device_schedule_callback(dev, ccwgroup_ungroup_callback);
+
+	return NOTIFY_OK;
+}
+
+
+/**
+ * ccwgroup_driver_register() - register a ccw group driver
+ * @cdriver: driver to be registered
+ *
+ * This function is mainly a wrapper around driver_register().
+ */
+int ccwgroup_driver_register(struct ccwgroup_driver *cdriver)
+{
+	/* register our new driver with the core */
+	cdriver->driver.bus = &ccwgroup_bus_type;
+	cdriver->driver.name = cdriver->name;
+	cdriver->driver.owner = cdriver->owner;
+
+	return driver_register(&cdriver->driver);
+}
+
+static int
+__ccwgroup_match_all(struct device *dev, void *data)
+{
+	return 1;
+}
+
+/**
+ * ccwgroup_driver_unregister() - deregister a ccw group driver
+ * @cdriver: driver to be deregistered
+ *
+ * This function is mainly a wrapper around driver_unregister().
+ */
+void ccwgroup_driver_unregister(struct ccwgroup_driver *cdriver)
+{
+	struct device *dev;
+
+	/* We don't want ccwgroup devices to live longer than their driver. */
+	get_driver(&cdriver->driver);
+	while ((dev = driver_find_device(&cdriver->driver, NULL, NULL,
+					 __ccwgroup_match_all))) {
+		struct ccwgroup_device *gdev = to_ccwgroupdev(dev);
+
+		mutex_lock(&gdev->reg_mutex);
+		__ccwgroup_remove_symlinks(gdev);
+		device_unregister(dev);
+		mutex_unlock(&gdev->reg_mutex);
+		put_device(dev);
+	}
+	put_driver(&cdriver->driver);
+	driver_unregister(&cdriver->driver);
+}
+
+/**
+ * ccwgroup_probe_ccwdev() - probe function for slave devices
+ * @cdev: ccw device to be probed
+ *
+ * This is a dummy probe function for ccw devices that are slave devices in
+ * a ccw group device.
+ * Returns:
+ *  always %0
+ */
+int ccwgroup_probe_ccwdev(struct ccw_device *cdev)
+{
+	return 0;
+}
+
+static struct ccwgroup_device *
+__ccwgroup_get_gdev_by_cdev(struct ccw_device *cdev)
+{
+	struct ccwgroup_device *gdev;
+
+	gdev = dev_get_drvdata(&cdev->dev);
+	if (gdev) {
+		if (get_device(&gdev->dev)) {
+			mutex_lock(&gdev->reg_mutex);
+			if (device_is_registered(&gdev->dev))
+				return gdev;
+			mutex_unlock(&gdev->reg_mutex);
+			put_device(&gdev->dev);
+		}
+		return NULL;
+	}
+	return NULL;
+}
+
+/**
+ * ccwgroup_remove_ccwdev() - remove function for slave devices
+ * @cdev: ccw device to be removed
+ *
+ * This is a remove function for ccw devices that are slave devices in a ccw
+ * group device. It sets the ccw device offline and also deregisters the
+ * embedding ccw group device.
+ */
+void ccwgroup_remove_ccwdev(struct ccw_device *cdev)
+{
+	struct ccwgroup_device *gdev;
+
+	/* Ignore offlining errors, device is gone anyway. */
+	ccw_device_set_offline(cdev);
+	/* If one of its devices is gone, the whole group is done for. */
+	gdev = __ccwgroup_get_gdev_by_cdev(cdev);
+	if (gdev) {
+		__ccwgroup_remove_symlinks(gdev);
+		device_unregister(&gdev->dev);
+		mutex_unlock(&gdev->reg_mutex);
+		put_device(&gdev->dev);
+	}
+}
+
+MODULE_LICENSE("GPL");
+EXPORT_SYMBOL(ccwgroup_driver_register);
+EXPORT_SYMBOL(ccwgroup_driver_unregister);
+EXPORT_SYMBOL(ccwgroup_probe_ccwdev);
+EXPORT_SYMBOL(ccwgroup_remove_ccwdev);

kernel/drivers/s390/cio/chp.c

@@ -0,0 +1,731 @@
+/*
+ *  drivers/s390/cio/chp.c
+ *
+ *    Copyright IBM Corp. 1999,2007
+ *    Author(s): Cornelia Huck (cornelia.huck@de.ibm.com)
+ *		 Arnd Bergmann (arndb@de.ibm.com)
+ *		 Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#include <linux/bug.h>
+#include <linux/workqueue.h>
+#include <linux/spinlock.h>
+#include <linux/init.h>
+#include <linux/jiffies.h>
+#include <linux/wait.h>
+#include <linux/mutex.h>
+#include <linux/errno.h>
+#include <asm/chpid.h>
+#include <asm/sclp.h>
+#include <asm/crw.h>
+
+#include "cio.h"
+#include "css.h"
+#include "ioasm.h"
+#include "cio_debug.h"
+#include "chp.h"
+
+#define to_channelpath(device) container_of(device, struct channel_path, dev)
+#define CHP_INFO_UPDATE_INTERVAL	1*HZ
+
+enum cfg_task_t {
+	cfg_none,
+	cfg_configure,
+	cfg_deconfigure
+};
+
+/* Map for pending configure tasks. */
+static enum cfg_task_t chp_cfg_task[__MAX_CSSID + 1][__MAX_CHPID + 1];
+static DEFINE_MUTEX(cfg_lock);
+static int cfg_busy;
+
+/* Map for channel-path status. */
+static struct sclp_chp_info chp_info;
+static DEFINE_MUTEX(info_lock);
+
+/* Time after which channel-path status may be outdated. */
+static unsigned long chp_info_expires;
+
+/* Workqueue to perform pending configure tasks. */
+static struct workqueue_struct *chp_wq;
+static struct work_struct cfg_work;
+
+/* Wait queue for configure completion events. */
+static wait_queue_head_t cfg_wait_queue;
+
+/* Return channel_path struct for given chpid. */
+static inline struct channel_path *chpid_to_chp(struct chp_id chpid)
+{
+	return channel_subsystems[chpid.cssid]->chps[chpid.id];
+}
+
+/* Set vary state for given chpid. */
+static void set_chp_logically_online(struct chp_id chpid, int onoff)
+{
+	chpid_to_chp(chpid)->state = onoff;
+}
+
+/* On succes return 0 if channel-path is varied offline, 1 if it is varied
+ * online. Return -ENODEV if channel-path is not registered. */
+int chp_get_status(struct chp_id chpid)
+{
+	return (chpid_to_chp(chpid) ? chpid_to_chp(chpid)->state : -ENODEV);
+}
+
+/**
+ * chp_get_sch_opm - return opm for subchannel
+ * @sch: subchannel
+ *
+ * Calculate and return the operational path mask (opm) based on the chpids
+ * used by the subchannel and the status of the associated channel-paths.
+ */
+u8 chp_get_sch_opm(struct subchannel *sch)
+{
+	struct chp_id chpid;
+	int opm;
+	int i;
+
+	opm = 0;
+	chp_id_init(&chpid);
+	for (i = 0; i < 8; i++) {
+		opm <<= 1;
+		chpid.id = sch->schib.pmcw.chpid[i];
+		if (chp_get_status(chpid) != 0)
+			opm |= 1;
+	}
+	return opm;
+}
+EXPORT_SYMBOL_GPL(chp_get_sch_opm);
+
+/**
+ * chp_is_registered - check if a channel-path is registered
+ * @chpid: channel-path ID
+ *
+ * Return non-zero if a channel-path with the given chpid is registered,
+ * zero otherwise.
+ */
+int chp_is_registered(struct chp_id chpid)
+{
+	return chpid_to_chp(chpid) != NULL;
+}
+
+/*
+ * Function: s390_vary_chpid
+ * Varies the specified chpid online or offline
+ */
+static int s390_vary_chpid(struct chp_id chpid, int on)
+{
+	char dbf_text[15];
+	int status;
+
+	sprintf(dbf_text, on?"varyon%x.%02x":"varyoff%x.%02x", chpid.cssid,
+		chpid.id);
+	CIO_TRACE_EVENT(2, dbf_text);
+
+	status = chp_get_status(chpid);
+	if (!on && !status)
+		return 0;
+
+	set_chp_logically_online(chpid, on);
+	chsc_chp_vary(chpid, on);
+	return 0;
+}
+
+/*
+ * Channel measurement related functions
+ */
+static ssize_t chp_measurement_chars_read(struct kobject *kobj,
+					  struct bin_attribute *bin_attr,
+					  char *buf, loff_t off, size_t count)
+{
+	struct channel_path *chp;
+	struct device *device;
+
+	device = container_of(kobj, struct device, kobj);
+	chp = to_channelpath(device);
+	if (!chp->cmg_chars)
+		return 0;
+
+	return memory_read_from_buffer(buf, count, &off,
+				chp->cmg_chars, sizeof(struct cmg_chars));
+}
+
+static struct bin_attribute chp_measurement_chars_attr = {
+	.attr = {
+		.name = "measurement_chars",
+		.mode = S_IRUSR,
+	},
+	.size = sizeof(struct cmg_chars),
+	.read = chp_measurement_chars_read,
+};
+
+static void chp_measurement_copy_block(struct cmg_entry *buf,
+				       struct channel_subsystem *css,
+				       struct chp_id chpid)
+{
+	void *area;
+	struct cmg_entry *entry, reference_buf;
+	int idx;
+
+	if (chpid.id < 128) {
+		area = css->cub_addr1;
+		idx = chpid.id;
+	} else {
+		area = css->cub_addr2;
+		idx = chpid.id - 128;
+	}
+	entry = area + (idx * sizeof(struct cmg_entry));
+	do {
+		memcpy(buf, entry, sizeof(*entry));
+		memcpy(&reference_buf, entry, sizeof(*entry));
+	} while (reference_buf.values[0] != buf->values[0]);
+}
+
+static ssize_t chp_measurement_read(struct kobject *kobj,
+				    struct bin_attribute *bin_attr,
+				    char *buf, loff_t off, size_t count)
+{
+	struct channel_path *chp;
+	struct channel_subsystem *css;
+	struct device *device;
+	unsigned int size;
+
+	device = container_of(kobj, struct device, kobj);
+	chp = to_channelpath(device);
+	css = to_css(chp->dev.parent);
+
+	size = sizeof(struct cmg_entry);
+
+	/* Only allow single reads. */
+	if (off || count < size)
+		return 0;
+	chp_measurement_copy_block((struct cmg_entry *)buf, css, chp->chpid);
+	count = size;
+	return count;
+}
+
+static struct bin_attribute chp_measurement_attr = {
+	.attr = {
+		.name = "measurement",
+		.mode = S_IRUSR,
+	},
+	.size = sizeof(struct cmg_entry),
+	.read = chp_measurement_read,
+};
+
+void chp_remove_cmg_attr(struct channel_path *chp)
+{
+	device_remove_bin_file(&chp->dev, &chp_measurement_chars_attr);
+	device_remove_bin_file(&chp->dev, &chp_measurement_attr);
+}
+
+int chp_add_cmg_attr(struct channel_path *chp)
+{
+	int ret;
+
+	ret = device_create_bin_file(&chp->dev, &chp_measurement_chars_attr);
+	if (ret)
+		return ret;
+	ret = device_create_bin_file(&chp->dev, &chp_measurement_attr);
+	if (ret)
+		device_remove_bin_file(&chp->dev, &chp_measurement_chars_attr);
+	return ret;
+}
+
+/*
+ * Files for the channel path entries.
+ */
+static ssize_t chp_status_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	struct channel_path *chp = to_channelpath(dev);
+
+	if (!chp)
+		return 0;
+	return (chp_get_status(chp->chpid) ? sprintf(buf, "online\n") :
+		sprintf(buf, "offline\n"));
+}
+
+static ssize_t chp_status_write(struct device *dev,
+				struct device_attribute *attr,
+				const char *buf, size_t count)
+{
+	struct channel_path *cp = to_channelpath(dev);
+	char cmd[10];
+	int num_args;
+	int error;
+
+	num_args = sscanf(buf, "%5s", cmd);
+	if (!num_args)
+		return count;
+
+	if (!strnicmp(cmd, "on", 2) || !strcmp(cmd, "1"))
+		error = s390_vary_chpid(cp->chpid, 1);
+	else if (!strnicmp(cmd, "off", 3) || !strcmp(cmd, "0"))
+		error = s390_vary_chpid(cp->chpid, 0);
+	else
+		error = -EINVAL;
+
+	return error < 0 ? error : count;
+
+}
+
+static DEVICE_ATTR(status, 0644, chp_status_show, chp_status_write);
+
+static ssize_t chp_configure_show(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	struct channel_path *cp;
+	int status;
+
+	cp = to_channelpath(dev);
+	status = chp_info_get_status(cp->chpid);
+	if (status < 0)
+		return status;
+
+	return snprintf(buf, PAGE_SIZE, "%d\n", status);
+}
+
+static int cfg_wait_idle(void);
+
+static ssize_t chp_configure_write(struct device *dev,
+				   struct device_attribute *attr,
+				   const char *buf, size_t count)
+{
+	struct channel_path *cp;
+	int val;
+	char delim;
+
+	if (sscanf(buf, "%d %c", &val, &delim) != 1)
+		return -EINVAL;
+	if (val != 0 && val != 1)
+		return -EINVAL;
+	cp = to_channelpath(dev);
+	chp_cfg_schedule(cp->chpid, val);
+	cfg_wait_idle();
+
+	return count;
+}
+
+static DEVICE_ATTR(configure, 0644, chp_configure_show, chp_configure_write);
+
+static ssize_t chp_type_show(struct device *dev, struct device_attribute *attr,
+			     char *buf)
+{
+	struct channel_path *chp = to_channelpath(dev);
+
+	if (!chp)
+		return 0;
+	return sprintf(buf, "%x\n", chp->desc.desc);
+}
+
+static DEVICE_ATTR(type, 0444, chp_type_show, NULL);
+
+static ssize_t chp_cmg_show(struct device *dev, struct device_attribute *attr,
+			    char *buf)
+{
+	struct channel_path *chp = to_channelpath(dev);
+
+	if (!chp)
+		return 0;
+	if (chp->cmg == -1) /* channel measurements not available */
+		return sprintf(buf, "unknown\n");
+	return sprintf(buf, "%x\n", chp->cmg);
+}
+
+static DEVICE_ATTR(cmg, 0444, chp_cmg_show, NULL);
+
+static ssize_t chp_shared_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	struct channel_path *chp = to_channelpath(dev);
+
+	if (!chp)
+		return 0;
+	if (chp->shared == -1) /* channel measurements not available */
+		return sprintf(buf, "unknown\n");
+	return sprintf(buf, "%x\n", chp->shared);
+}
+
+static DEVICE_ATTR(shared, 0444, chp_shared_show, NULL);
+
+static struct attribute *chp_attrs[] = {
+	&dev_attr_status.attr,
+	&dev_attr_configure.attr,
+	&dev_attr_type.attr,
+	&dev_attr_cmg.attr,
+	&dev_attr_shared.attr,
+	NULL,
+};
+
+static struct attribute_group chp_attr_group = {
+	.attrs = chp_attrs,
+};
+
+static void chp_release(struct device *dev)
+{
+	struct channel_path *cp;
+
+	cp = to_channelpath(dev);
+	kfree(cp);
+}
+
+/**
+ * chp_new - register a new channel-path
+ * @chpid - channel-path ID
+ *
+ * Create and register data structure representing new channel-path. Return
+ * zero on success, non-zero otherwise.
+ */
+int chp_new(struct chp_id chpid)
+{
+	struct channel_path *chp;
+	int ret;
+
+	if (chp_is_registered(chpid))
+		return 0;
+	chp = kzalloc(sizeof(struct channel_path), GFP_KERNEL);
+	if (!chp)
+		return -ENOMEM;
+
+	/* fill in status, etc. */
+	chp->chpid = chpid;
+	chp->state = 1;
+	chp->dev.parent = &channel_subsystems[chpid.cssid]->device;
+	chp->dev.release = chp_release;
+
+	/* Obtain channel path description and fill it in. */
+	ret = chsc_determine_base_channel_path_desc(chpid, &chp->desc);
+	if (ret)
+		goto out_free;
+	if ((chp->desc.flags & 0x80) == 0) {
+		ret = -ENODEV;
+		goto out_free;
+	}
+	/* Get channel-measurement characteristics. */
+	if (css_chsc_characteristics.scmc && css_chsc_characteristics.secm) {
+		ret = chsc_get_channel_measurement_chars(chp);
+		if (ret)
+			goto out_free;
+	} else {
+		chp->cmg = -1;
+	}
+	dev_set_name(&chp->dev, "chp%x.%02x", chpid.cssid, chpid.id);
+
+	/* make it known to the system */
+	ret = device_register(&chp->dev);
+	if (ret) {
+		CIO_MSG_EVENT(0, "Could not register chp%x.%02x: %d\n",
+			      chpid.cssid, chpid.id, ret);
+		put_device(&chp->dev);
+		goto out;
+	}
+	ret = sysfs_create_group(&chp->dev.kobj, &chp_attr_group);
+	if (ret) {
+		device_unregister(&chp->dev);
+		goto out;
+	}
+	mutex_lock(&channel_subsystems[chpid.cssid]->mutex);
+	if (channel_subsystems[chpid.cssid]->cm_enabled) {
+		ret = chp_add_cmg_attr(chp);
+		if (ret) {
+			sysfs_remove_group(&chp->dev.kobj, &chp_attr_group);
+			device_unregister(&chp->dev);
+			mutex_unlock(&channel_subsystems[chpid.cssid]->mutex);
+			goto out;
+		}
+	}
+	channel_subsystems[chpid.cssid]->chps[chpid.id] = chp;
+	mutex_unlock(&channel_subsystems[chpid.cssid]->mutex);
+	goto out;
+out_free:
+	kfree(chp);
+out:
+	return ret;
+}
+
+/**
+ * chp_get_chp_desc - return newly allocated channel-path description
+ * @chpid: channel-path ID
+ *
+ * On success return a newly allocated copy of the channel-path description
+ * data associated with the given channel-path ID. Return %NULL on error.
+ */
+void *chp_get_chp_desc(struct chp_id chpid)
+{
+	struct channel_path *chp;
+	struct channel_path_desc *desc;
+
+	chp = chpid_to_chp(chpid);
+	if (!chp)
+		return NULL;
+	desc = kmalloc(sizeof(struct channel_path_desc), GFP_KERNEL);
+	if (!desc)
+		return NULL;
+	memcpy(desc, &chp->desc, sizeof(struct channel_path_desc));
+	return desc;
+}
+
+/**
+ * chp_process_crw - process channel-path status change
+ * @crw0: channel report-word to handler
+ * @crw1: second channel-report word (always NULL)
+ * @overflow: crw overflow indication
+ *
+ * Handle channel-report-words indicating that the status of a channel-path
+ * has changed.
+ */
+static void chp_process_crw(struct crw *crw0, struct crw *crw1,
+			    int overflow)
+{
+	struct chp_id chpid;
+
+	if (overflow) {
+		css_schedule_eval_all();
+		return;
+	}
+	CIO_CRW_EVENT(2, "CRW reports slct=%d, oflw=%d, "
+		      "chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
+		      crw0->slct, crw0->oflw, crw0->chn, crw0->rsc, crw0->anc,
+		      crw0->erc, crw0->rsid);
+	/*
+	 * Check for solicited machine checks. These are
+	 * created by reset channel path and need not be
+	 * handled here.
+	 */
+	if (crw0->slct) {
+		CIO_CRW_EVENT(2, "solicited machine check for "
+			      "channel path %02X\n", crw0->rsid);
+		return;
+	}
+	chp_id_init(&chpid);
+	chpid.id = crw0->rsid;
+	switch (crw0->erc) {
+	case CRW_ERC_IPARM: /* Path has come. */
+		if (!chp_is_registered(chpid))
+			chp_new(chpid);
+		chsc_chp_online(chpid);
+		break;
+	case CRW_ERC_PERRI: /* Path has gone. */
+	case CRW_ERC_PERRN:
+		chsc_chp_offline(chpid);
+		break;
+	default:
+		CIO_CRW_EVENT(2, "Don't know how to handle erc=%x\n",
+			      crw0->erc);
+	}
+}
+
+int chp_ssd_get_mask(struct chsc_ssd_info *ssd, struct chp_link *link)
+{
+	int i;
+	int mask;
+
+	for (i = 0; i < 8; i++) {
+		mask = 0x80 >> i;
+		if (!(ssd->path_mask & mask))
+			continue;
+		if (!chp_id_is_equal(&ssd->chpid[i], &link->chpid))
+			continue;
+		if ((ssd->fla_valid_mask & mask) &&
+		    ((ssd->fla[i] & link->fla_mask) != link->fla))
+			continue;
+		return mask;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(chp_ssd_get_mask);
+
+static inline int info_bit_num(struct chp_id id)
+{
+	return id.id + id.cssid * (__MAX_CHPID + 1);
+}
+
+/* Force chp_info refresh on next call to info_validate(). */
+static void info_expire(void)
+{
+	mutex_lock(&info_lock);
+	chp_info_expires = jiffies - 1;
+	mutex_unlock(&info_lock);
+}
+
+/* Ensure that chp_info is up-to-date. */
+static int info_update(void)
+{
+	int rc;
+
+	mutex_lock(&info_lock);
+	rc = 0;
+	if (time_after(jiffies, chp_info_expires)) {
+		/* Data is too old, update. */
+		rc = sclp_chp_read_info(&chp_info);
+		chp_info_expires = jiffies + CHP_INFO_UPDATE_INTERVAL ;
+	}
+	mutex_unlock(&info_lock);
+
+	return rc;
+}
+
+/**
+ * chp_info_get_status - retrieve configure status of a channel-path
+ * @chpid: channel-path ID
+ *
+ * On success, return 0 for standby, 1 for configured, 2 for reserved,
+ * 3 for not recognized. Return negative error code on error.
+ */
+int chp_info_get_status(struct chp_id chpid)
+{
+	int rc;
+	int bit;
+
+	rc = info_update();
+	if (rc)
+		return rc;
+
+	bit = info_bit_num(chpid);
+	mutex_lock(&info_lock);
+	if (!chp_test_bit(chp_info.recognized, bit))
+		rc = CHP_STATUS_NOT_RECOGNIZED;
+	else if (chp_test_bit(chp_info.configured, bit))
+		rc = CHP_STATUS_CONFIGURED;
+	else if (chp_test_bit(chp_info.standby, bit))
+		rc = CHP_STATUS_STANDBY;
+	else
+		rc = CHP_STATUS_RESERVED;
+	mutex_unlock(&info_lock);
+
+	return rc;
+}
+
+/* Return configure task for chpid. */
+static enum cfg_task_t cfg_get_task(struct chp_id chpid)
+{
+	return chp_cfg_task[chpid.cssid][chpid.id];
+}
+
+/* Set configure task for chpid. */
+static void cfg_set_task(struct chp_id chpid, enum cfg_task_t cfg)
+{
+	chp_cfg_task[chpid.cssid][chpid.id] = cfg;
+}
+
+/* Perform one configure/deconfigure request. Reschedule work function until
+ * last request. */
+static void cfg_func(struct work_struct *work)
+{
+	struct chp_id chpid;
+	enum cfg_task_t t;
+	int rc;
+
+	mutex_lock(&cfg_lock);
+	t = cfg_none;
+	chp_id_for_each(&chpid) {
+		t = cfg_get_task(chpid);
+		if (t != cfg_none) {
+			cfg_set_task(chpid, cfg_none);
+			break;
+		}
+	}
+	mutex_unlock(&cfg_lock);
+
+	switch (t) {
+	case cfg_configure:
+		rc = sclp_chp_configure(chpid);
+		if (rc)
+			CIO_MSG_EVENT(2, "chp: sclp_chp_configure(%x.%02x)="
+				      "%d\n", chpid.cssid, chpid.id, rc);
+		else {
+			info_expire();
+			chsc_chp_online(chpid);
+		}
+		break;
+	case cfg_deconfigure:
+		rc = sclp_chp_deconfigure(chpid);
+		if (rc)
+			CIO_MSG_EVENT(2, "chp: sclp_chp_deconfigure(%x.%02x)="
+				      "%d\n", chpid.cssid, chpid.id, rc);
+		else {
+			info_expire();
+			chsc_chp_offline(chpid);
+		}
+		break;
+	case cfg_none:
+		/* Get updated information after last change. */
+		info_update();
+		mutex_lock(&cfg_lock);
+		cfg_busy = 0;
+		mutex_unlock(&cfg_lock);
+		wake_up_interruptible(&cfg_wait_queue);
+		return;
+	}
+	queue_work(chp_wq, &cfg_work);
+}
+
+/**
+ * chp_cfg_schedule - schedule chpid configuration request
+ * @chpid - channel-path ID
+ * @configure - Non-zero for configure, zero for deconfigure
+ *
+ * Schedule a channel-path configuration/deconfiguration request.
+ */
+void chp_cfg_schedule(struct chp_id chpid, int configure)
+{
+	CIO_MSG_EVENT(2, "chp_cfg_sched%x.%02x=%d\n", chpid.cssid, chpid.id,
+		      configure);
+	mutex_lock(&cfg_lock);
+	cfg_set_task(chpid, configure ? cfg_configure : cfg_deconfigure);
+	cfg_busy = 1;
+	mutex_unlock(&cfg_lock);
+	queue_work(chp_wq, &cfg_work);
+}
+
+/**
+ * chp_cfg_cancel_deconfigure - cancel chpid deconfiguration request
+ * @chpid - channel-path ID
+ *
+ * Cancel an active channel-path deconfiguration request if it has not yet
+ * been performed.
+ */
+void chp_cfg_cancel_deconfigure(struct chp_id chpid)
+{
+	CIO_MSG_EVENT(2, "chp_cfg_cancel:%x.%02x\n", chpid.cssid, chpid.id);
+	mutex_lock(&cfg_lock);
+	if (cfg_get_task(chpid) == cfg_deconfigure)
+		cfg_set_task(chpid, cfg_none);
+	mutex_unlock(&cfg_lock);
+}
+
+static int cfg_wait_idle(void)
+{
+	if (wait_event_interruptible(cfg_wait_queue, !cfg_busy))
+		return -ERESTARTSYS;
+	return 0;
+}
+
+static int __init chp_init(void)
+{
+	struct chp_id chpid;
+	int ret;
+
+	ret = crw_register_handler(CRW_RSC_CPATH, chp_process_crw);
+	if (ret)
+		return ret;
+	chp_wq = create_singlethread_workqueue("cio_chp");
+	if (!chp_wq) {
+		crw_unregister_handler(CRW_RSC_CPATH);
+		return -ENOMEM;
+	}
+	INIT_WORK(&cfg_work, cfg_func);
+	init_waitqueue_head(&cfg_wait_queue);
+	if (info_update())
+		return 0;
+	/* Register available channel-paths. */
+	chp_id_for_each(&chpid) {
+		if (chp_info_get_status(chpid) != CHP_STATUS_NOT_RECOGNIZED)
+			chp_new(chpid);
+	}
+
+	return 0;
+}
+
+subsys_initcall(chp_init);

kernel/drivers/s390/cio/chp.h

@@ -0,0 +1,64 @@
+/*
+ *  drivers/s390/cio/chp.h
+ *
+ *    Copyright IBM Corp. 2007
+ *    Author(s): Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#ifndef S390_CHP_H
+#define S390_CHP_H S390_CHP_H
+
+#include <linux/types.h>
+#include <linux/device.h>
+#include <asm/chpid.h>
+#include "chsc.h"
+#include "css.h"
+
+#define CHP_STATUS_STANDBY		0
+#define CHP_STATUS_CONFIGURED		1
+#define CHP_STATUS_RESERVED		2
+#define CHP_STATUS_NOT_RECOGNIZED	3
+
+#define CHP_ONLINE 0
+#define CHP_OFFLINE 1
+#define CHP_VARY_ON 2
+#define CHP_VARY_OFF 3
+
+struct chp_link {
+	struct chp_id chpid;
+	u32 fla_mask;
+	u16 fla;
+};
+
+static inline int chp_test_bit(u8 *bitmap, int num)
+{
+	int byte = num >> 3;
+	int mask = 128 >> (num & 7);
+
+	return (bitmap[byte] & mask) ? 1 : 0;
+}
+
+
+struct channel_path {
+	struct chp_id chpid;
+	int state;
+	struct channel_path_desc desc;
+	/* Channel-measurement related stuff: */
+	int cmg;
+	int shared;
+	void *cmg_chars;
+	struct device dev;
+};
+
+int chp_get_status(struct chp_id chpid);
+u8 chp_get_sch_opm(struct subchannel *sch);
+int chp_is_registered(struct chp_id chpid);
+void *chp_get_chp_desc(struct chp_id chpid);
+void chp_remove_cmg_attr(struct channel_path *chp);
+int chp_add_cmg_attr(struct channel_path *chp);
+int chp_new(struct chp_id chpid);
+void chp_cfg_schedule(struct chp_id chpid, int configure);
+void chp_cfg_cancel_deconfigure(struct chp_id chpid);
+int chp_info_get_status(struct chp_id chpid);
+int chp_ssd_get_mask(struct chsc_ssd_info *, struct chp_link *);
+#endif /* S390_CHP_H */

kernel/drivers/s390/cio/chsc.c

@@ -0,0 +1,977 @@
+/*
+ *  drivers/s390/cio/chsc.c
+ *   S/390 common I/O routines -- channel subsystem call
+ *
+ *    Copyright IBM Corp. 1999,2008
+ *    Author(s): Ingo Adlung (adlung@de.ibm.com)
+ *		 Cornelia Huck (cornelia.huck@de.ibm.com)
+ *		 Arnd Bergmann (arndb@de.ibm.com)
+ */
+
+#define KMSG_COMPONENT "cio"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <linux/device.h>
+
+#include <asm/cio.h>
+#include <asm/chpid.h>
+#include <asm/chsc.h>
+#include <asm/crw.h>
+
+#include "css.h"
+#include "cio.h"
+#include "cio_debug.h"
+#include "ioasm.h"
+#include "chp.h"
+#include "chsc.h"
+
+static void *sei_page;
+
+/**
+ * chsc_error_from_response() - convert a chsc response to an error
+ * @response: chsc response code
+ *
+ * Returns an appropriate Linux error code for @response.
+ */
+int chsc_error_from_response(int response)
+{
+	switch (response) {
+	case 0x0001:
+		return 0;
+	case 0x0002:
+	case 0x0003:
+	case 0x0006:
+	case 0x0007:
+	case 0x0008:
+	case 0x000a:
+		return -EINVAL;
+	case 0x0004:
+		return -EOPNOTSUPP;
+	default:
+		return -EIO;
+	}
+}
+EXPORT_SYMBOL_GPL(chsc_error_from_response);
+
+struct chsc_ssd_area {
+	struct chsc_header request;
+	u16 :10;
+	u16 ssid:2;
+	u16 :4;
+	u16 f_sch;	  /* first subchannel */
+	u16 :16;
+	u16 l_sch;	  /* last subchannel */
+	u32 :32;
+	struct chsc_header response;
+	u32 :32;
+	u8 sch_valid : 1;
+	u8 dev_valid : 1;
+	u8 st	     : 3; /* subchannel type */
+	u8 zeroes    : 3;
+	u8  unit_addr;	  /* unit address */
+	u16 devno;	  /* device number */
+	u8 path_mask;
+	u8 fla_valid_mask;
+	u16 sch;	  /* subchannel */
+	u8 chpid[8];	  /* chpids 0-7 */
+	u16 fla[8];	  /* full link addresses 0-7 */
+} __attribute__ ((packed));
+
+int chsc_get_ssd_info(struct subchannel_id schid, struct chsc_ssd_info *ssd)
+{
+	unsigned long page;
+	struct chsc_ssd_area *ssd_area;
+	int ccode;
+	int ret;
+	int i;
+	int mask;
+
+	page = get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!page)
+		return -ENOMEM;
+	ssd_area = (struct chsc_ssd_area *) page;
+	ssd_area->request.length = 0x0010;
+	ssd_area->request.code = 0x0004;
+	ssd_area->ssid = schid.ssid;
+	ssd_area->f_sch = schid.sch_no;
+	ssd_area->l_sch = schid.sch_no;
+
+	ccode = chsc(ssd_area);
+	/* Check response. */
+	if (ccode > 0) {
+		ret = (ccode == 3) ? -ENODEV : -EBUSY;
+		goto out_free;
+	}
+	ret = chsc_error_from_response(ssd_area->response.code);
+	if (ret != 0) {
+		CIO_MSG_EVENT(2, "chsc: ssd failed for 0.%x.%04x (rc=%04x)\n",
+			      schid.ssid, schid.sch_no,
+			      ssd_area->response.code);
+		goto out_free;
+	}
+	if (!ssd_area->sch_valid) {
+		ret = -ENODEV;
+		goto out_free;
+	}
+	/* Copy data */
+	ret = 0;
+	memset(ssd, 0, sizeof(struct chsc_ssd_info));
+	if ((ssd_area->st != SUBCHANNEL_TYPE_IO) &&
+	    (ssd_area->st != SUBCHANNEL_TYPE_MSG))
+		goto out_free;
+	ssd->path_mask = ssd_area->path_mask;
+	ssd->fla_valid_mask = ssd_area->fla_valid_mask;
+	for (i = 0; i < 8; i++) {
+		mask = 0x80 >> i;
+		if (ssd_area->path_mask & mask) {
+			chp_id_init(&ssd->chpid[i]);
+			ssd->chpid[i].id = ssd_area->chpid[i];
+		}
+		if (ssd_area->fla_valid_mask & mask)
+			ssd->fla[i] = ssd_area->fla[i];
+	}
+out_free:
+	free_page(page);
+	return ret;
+}
+
+static int s390_subchannel_remove_chpid(struct subchannel *sch, void *data)
+{
+	spin_lock_irq(sch->lock);
+	if (sch->driver && sch->driver->chp_event)
+		if (sch->driver->chp_event(sch, data, CHP_OFFLINE) != 0)
+			goto out_unreg;
+	spin_unlock_irq(sch->lock);
+	return 0;
+
+out_unreg:
+	sch->lpm = 0;
+	spin_unlock_irq(sch->lock);
+	css_schedule_eval(sch->schid);
+	return 0;
+}
+
+void chsc_chp_offline(struct chp_id chpid)
+{
+	char dbf_txt[15];
+	struct chp_link link;
+
+	sprintf(dbf_txt, "chpr%x.%02x", chpid.cssid, chpid.id);
+	CIO_TRACE_EVENT(2, dbf_txt);
+
+	if (chp_get_status(chpid) <= 0)
+		return;
+	memset(&link, 0, sizeof(struct chp_link));
+	link.chpid = chpid;
+	/* Wait until previous actions have settled. */
+	css_wait_for_slow_path();
+	for_each_subchannel_staged(s390_subchannel_remove_chpid, NULL, &link);
+}
+
+static int s390_process_res_acc_new_sch(struct subchannel_id schid, void *data)
+{
+	struct schib schib;
+	/*
+	 * We don't know the device yet, but since a path
+	 * may be available now to the device we'll have
+	 * to do recognition again.
+	 * Since we don't have any idea about which chpid
+	 * that beast may be on we'll have to do a stsch
+	 * on all devices, grr...
+	 */
+	if (stsch_err(schid, &schib))
+		/* We're through */
+		return -ENXIO;
+
+	/* Put it on the slow path. */
+	css_schedule_eval(schid);
+	return 0;
+}
+
+static int __s390_process_res_acc(struct subchannel *sch, void *data)
+{
+	spin_lock_irq(sch->lock);
+	if (sch->driver && sch->driver->chp_event)
+		sch->driver->chp_event(sch, data, CHP_ONLINE);
+	spin_unlock_irq(sch->lock);
+
+	return 0;
+}
+
+static void s390_process_res_acc(struct chp_link *link)
+{
+	char dbf_txt[15];
+
+	sprintf(dbf_txt, "accpr%x.%02x", link->chpid.cssid,
+		link->chpid.id);
+	CIO_TRACE_EVENT( 2, dbf_txt);
+	if (link->fla != 0) {
+		sprintf(dbf_txt, "fla%x", link->fla);
+		CIO_TRACE_EVENT( 2, dbf_txt);
+	}
+	/* Wait until previous actions have settled. */
+	css_wait_for_slow_path();
+	/*
+	 * I/O resources may have become accessible.
+	 * Scan through all subchannels that may be concerned and
+	 * do a validation on those.
+	 * The more information we have (info), the less scanning
+	 * will we have to do.
+	 */
+	for_each_subchannel_staged(__s390_process_res_acc,
+				   s390_process_res_acc_new_sch, link);
+}
+
+static int
+__get_chpid_from_lir(void *data)
+{
+	struct lir {
+		u8  iq;
+		u8  ic;
+		u16 sci;
+		/* incident-node descriptor */
+		u32 indesc[28];
+		/* attached-node descriptor */
+		u32 andesc[28];
+		/* incident-specific information */
+		u32 isinfo[28];
+	} __attribute__ ((packed)) *lir;
+
+	lir = data;
+	if (!(lir->iq&0x80))
+		/* NULL link incident record */
+		return -EINVAL;
+	if (!(lir->indesc[0]&0xc0000000))
+		/* node descriptor not valid */
+		return -EINVAL;
+	if (!(lir->indesc[0]&0x10000000))
+		/* don't handle device-type nodes - FIXME */
+		return -EINVAL;
+	/* Byte 3 contains the chpid. Could also be CTCA, but we don't care */
+
+	return (u16) (lir->indesc[0]&0x000000ff);
+}
+
+struct chsc_sei_area {
+	struct chsc_header request;
+	u32 reserved1;
+	u32 reserved2;
+	u32 reserved3;
+	struct chsc_header response;
+	u32 reserved4;
+	u8  flags;
+	u8  vf;		/* validity flags */
+	u8  rs;		/* reporting source */
+	u8  cc;		/* content code */
+	u16 fla;	/* full link address */
+	u16 rsid;	/* reporting source id */
+	u32 reserved5;
+	u32 reserved6;
+	u8 ccdf[4096 - 16 - 24];	/* content-code dependent field */
+	/* ccdf has to be big enough for a link-incident record */
+} __attribute__ ((packed));
+
+static void chsc_process_sei_link_incident(struct chsc_sei_area *sei_area)
+{
+	struct chp_id chpid;
+	int id;
+
+	CIO_CRW_EVENT(4, "chsc: link incident (rs=%02x, rs_id=%04x)\n",
+		      sei_area->rs, sei_area->rsid);
+	if (sei_area->rs != 4)
+		return;
+	id = __get_chpid_from_lir(sei_area->ccdf);
+	if (id < 0)
+		CIO_CRW_EVENT(4, "chsc: link incident - invalid LIR\n");
+	else {
+		chp_id_init(&chpid);
+		chpid.id = id;
+		chsc_chp_offline(chpid);
+	}
+}
+
+static void chsc_process_sei_res_acc(struct chsc_sei_area *sei_area)
+{
+	struct chp_link link;
+	struct chp_id chpid;
+	int status;
+
+	CIO_CRW_EVENT(4, "chsc: resource accessibility event (rs=%02x, "
+		      "rs_id=%04x)\n", sei_area->rs, sei_area->rsid);
+	if (sei_area->rs != 4)
+		return;
+	chp_id_init(&chpid);
+	chpid.id = sei_area->rsid;
+	/* allocate a new channel path structure, if needed */
+	status = chp_get_status(chpid);
+	if (status < 0)
+		chp_new(chpid);
+	else if (!status)
+		return;
+	memset(&link, 0, sizeof(struct chp_link));
+	link.chpid = chpid;
+	if ((sei_area->vf & 0xc0) != 0) {
+		link.fla = sei_area->fla;
+		if ((sei_area->vf & 0xc0) == 0xc0)
+			/* full link address */
+			link.fla_mask = 0xffff;
+		else
+			/* link address */
+			link.fla_mask = 0xff00;
+	}
+	s390_process_res_acc(&link);
+}
+
+struct chp_config_data {
+	u8 map[32];
+	u8 op;
+	u8 pc;
+};
+
+static void chsc_process_sei_chp_config(struct chsc_sei_area *sei_area)
+{
+	struct chp_config_data *data;
+	struct chp_id chpid;
+	int num;
+	char *events[3] = {"configure", "deconfigure", "cancel deconfigure"};
+
+	CIO_CRW_EVENT(4, "chsc: channel-path-configuration notification\n");
+	if (sei_area->rs != 0)
+		return;
+	data = (struct chp_config_data *) &(sei_area->ccdf);
+	chp_id_init(&chpid);
+	for (num = 0; num <= __MAX_CHPID; num++) {
+		if (!chp_test_bit(data->map, num))
+			continue;
+		chpid.id = num;
+		pr_notice("Processing %s for channel path %x.%02x\n",
+			  events[data->op], chpid.cssid, chpid.id);
+		switch (data->op) {
+		case 0:
+			chp_cfg_schedule(chpid, 1);
+			break;
+		case 1:
+			chp_cfg_schedule(chpid, 0);
+			break;
+		case 2:
+			chp_cfg_cancel_deconfigure(chpid);
+			break;
+		}
+	}
+}
+
+static void chsc_process_sei(struct chsc_sei_area *sei_area)
+{
+	/* Check if we might have lost some information. */
+	if (sei_area->flags & 0x40) {
+		CIO_CRW_EVENT(2, "chsc: event overflow\n");
+		css_schedule_eval_all();
+	}
+	/* which kind of information was stored? */
+	switch (sei_area->cc) {
+	case 1: /* link incident*/
+		chsc_process_sei_link_incident(sei_area);
+		break;
+	case 2: /* i/o resource accessibiliy */
+		chsc_process_sei_res_acc(sei_area);
+		break;
+	case 8: /* channel-path-configuration notification */
+		chsc_process_sei_chp_config(sei_area);
+		break;
+	default: /* other stuff */
+		CIO_CRW_EVENT(4, "chsc: unhandled sei content code %d\n",
+			      sei_area->cc);
+		break;
+	}
+}
+
+static void chsc_process_crw(struct crw *crw0, struct crw *crw1, int overflow)
+{
+	struct chsc_sei_area *sei_area;
+
+	if (overflow) {
+		css_schedule_eval_all();
+		return;
+	}
+	CIO_CRW_EVENT(2, "CRW reports slct=%d, oflw=%d, "
+		      "chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
+		      crw0->slct, crw0->oflw, crw0->chn, crw0->rsc, crw0->anc,
+		      crw0->erc, crw0->rsid);
+	if (!sei_page)
+		return;
+	/* Access to sei_page is serialized through machine check handler
+	 * thread, so no need for locking. */
+	sei_area = sei_page;
+
+	CIO_TRACE_EVENT(2, "prcss");
+	do {
+		memset(sei_area, 0, sizeof(*sei_area));
+		sei_area->request.length = 0x0010;
+		sei_area->request.code = 0x000e;
+		if (chsc(sei_area))
+			break;
+
+		if (sei_area->response.code == 0x0001) {
+			CIO_CRW_EVENT(4, "chsc: sei successful\n");
+			chsc_process_sei(sei_area);
+		} else {
+			CIO_CRW_EVENT(2, "chsc: sei failed (rc=%04x)\n",
+				      sei_area->response.code);
+			break;
+		}
+	} while (sei_area->flags & 0x80);
+}
+
+void chsc_chp_online(struct chp_id chpid)
+{
+	char dbf_txt[15];
+	struct chp_link link;
+
+	sprintf(dbf_txt, "cadd%x.%02x", chpid.cssid, chpid.id);
+	CIO_TRACE_EVENT(2, dbf_txt);
+
+	if (chp_get_status(chpid) != 0) {
+		memset(&link, 0, sizeof(struct chp_link));
+		link.chpid = chpid;
+		/* Wait until previous actions have settled. */
+		css_wait_for_slow_path();
+		for_each_subchannel_staged(__s390_process_res_acc, NULL,
+					   &link);
+	}
+}
+
+static void __s390_subchannel_vary_chpid(struct subchannel *sch,
+					 struct chp_id chpid, int on)
+{
+	unsigned long flags;
+	struct chp_link link;
+
+	memset(&link, 0, sizeof(struct chp_link));
+	link.chpid = chpid;
+	spin_lock_irqsave(sch->lock, flags);
+	if (sch->driver && sch->driver->chp_event)
+		sch->driver->chp_event(sch, &link,
+				       on ? CHP_VARY_ON : CHP_VARY_OFF);
+	spin_unlock_irqrestore(sch->lock, flags);
+}
+
+static int s390_subchannel_vary_chpid_off(struct subchannel *sch, void *data)
+{
+	struct chp_id *chpid = data;
+
+	__s390_subchannel_vary_chpid(sch, *chpid, 0);
+	return 0;
+}
+
+static int s390_subchannel_vary_chpid_on(struct subchannel *sch, void *data)
+{
+	struct chp_id *chpid = data;
+
+	__s390_subchannel_vary_chpid(sch, *chpid, 1);
+	return 0;
+}
+
+static int
+__s390_vary_chpid_on(struct subchannel_id schid, void *data)
+{
+	struct schib schib;
+
+	if (stsch_err(schid, &schib))
+		/* We're through */
+		return -ENXIO;
+	/* Put it on the slow path. */
+	css_schedule_eval(schid);
+	return 0;
+}
+
+/**
+ * chsc_chp_vary - propagate channel-path vary operation to subchannels
+ * @chpid: channl-path ID
+ * @on: non-zero for vary online, zero for vary offline
+ */
+int chsc_chp_vary(struct chp_id chpid, int on)
+{
+	struct chp_link link;
+
+	memset(&link, 0, sizeof(struct chp_link));
+	link.chpid = chpid;
+	/* Wait until previous actions have settled. */
+	css_wait_for_slow_path();
+	/*
+	 * Redo PathVerification on the devices the chpid connects to
+	 */
+
+	if (on)
+		for_each_subchannel_staged(s390_subchannel_vary_chpid_on,
+					   __s390_vary_chpid_on, &link);
+	else
+		for_each_subchannel_staged(s390_subchannel_vary_chpid_off,
+					   NULL, &link);
+
+	return 0;
+}
+
+static void
+chsc_remove_cmg_attr(struct channel_subsystem *css)
+{
+	int i;
+
+	for (i = 0; i <= __MAX_CHPID; i++) {
+		if (!css->chps[i])
+			continue;
+		chp_remove_cmg_attr(css->chps[i]);
+	}
+}
+
+static int
+chsc_add_cmg_attr(struct channel_subsystem *css)
+{
+	int i, ret;
+
+	ret = 0;
+	for (i = 0; i <= __MAX_CHPID; i++) {
+		if (!css->chps[i])
+			continue;
+		ret = chp_add_cmg_attr(css->chps[i]);
+		if (ret)
+			goto cleanup;
+	}
+	return ret;
+cleanup:
+	for (--i; i >= 0; i--) {
+		if (!css->chps[i])
+			continue;
+		chp_remove_cmg_attr(css->chps[i]);
+	}
+	return ret;
+}
+
+int __chsc_do_secm(struct channel_subsystem *css, int enable, void *page)
+{
+	struct {
+		struct chsc_header request;
+		u32 operation_code : 2;
+		u32 : 30;
+		u32 key : 4;
+		u32 : 28;
+		u32 zeroes1;
+		u32 cub_addr1;
+		u32 zeroes2;
+		u32 cub_addr2;
+		u32 reserved[13];
+		struct chsc_header response;
+		u32 status : 8;
+		u32 : 4;
+		u32 fmt : 4;
+		u32 : 16;
+	} __attribute__ ((packed)) *secm_area;
+	int ret, ccode;
+
+	secm_area = page;
+	secm_area->request.length = 0x0050;
+	secm_area->request.code = 0x0016;
+
+	secm_area->key = PAGE_DEFAULT_KEY;
+	secm_area->cub_addr1 = (u64)(unsigned long)css->cub_addr1;
+	secm_area->cub_addr2 = (u64)(unsigned long)css->cub_addr2;
+
+	secm_area->operation_code = enable ? 0 : 1;
+
+	ccode = chsc(secm_area);
+	if (ccode > 0)
+		return (ccode == 3) ? -ENODEV : -EBUSY;
+
+	switch (secm_area->response.code) {
+	case 0x0102:
+	case 0x0103:
+		ret = -EINVAL;
+		break;
+	default:
+		ret = chsc_error_from_response(secm_area->response.code);
+	}
+	if (ret != 0)
+		CIO_CRW_EVENT(2, "chsc: secm failed (rc=%04x)\n",
+			      secm_area->response.code);
+	return ret;
+}
+
+int
+chsc_secm(struct channel_subsystem *css, int enable)
+{
+	void  *secm_area;
+	int ret;
+
+	secm_area = (void *)get_zeroed_page(GFP_KERNEL |  GFP_DMA);
+	if (!secm_area)
+		return -ENOMEM;
+
+	if (enable && !css->cm_enabled) {
+		css->cub_addr1 = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		css->cub_addr2 = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+		if (!css->cub_addr1 || !css->cub_addr2) {
+			free_page((unsigned long)css->cub_addr1);
+			free_page((unsigned long)css->cub_addr2);
+			free_page((unsigned long)secm_area);
+			return -ENOMEM;
+		}
+	}
+	ret = __chsc_do_secm(css, enable, secm_area);
+	if (!ret) {
+		css->cm_enabled = enable;
+		if (css->cm_enabled) {
+			ret = chsc_add_cmg_attr(css);
+			if (ret) {
+				memset(secm_area, 0, PAGE_SIZE);
+				__chsc_do_secm(css, 0, secm_area);
+				css->cm_enabled = 0;
+			}
+		} else
+			chsc_remove_cmg_attr(css);
+	}
+	if (!css->cm_enabled) {
+		free_page((unsigned long)css->cub_addr1);
+		free_page((unsigned long)css->cub_addr2);
+	}
+	free_page((unsigned long)secm_area);
+	return ret;
+}
+
+int chsc_determine_channel_path_desc(struct chp_id chpid, int fmt, int rfmt,
+				     int c, int m,
+				     struct chsc_response_struct *resp)
+{
+	int ccode, ret;
+
+	struct {
+		struct chsc_header request;
+		u32 : 2;
+		u32 m : 1;
+		u32 c : 1;
+		u32 fmt : 4;
+		u32 cssid : 8;
+		u32 : 4;
+		u32 rfmt : 4;
+		u32 first_chpid : 8;
+		u32 : 24;
+		u32 last_chpid : 8;
+		u32 zeroes1;
+		struct chsc_header response;
+		u8 data[PAGE_SIZE - 20];
+	} __attribute__ ((packed)) *scpd_area;
+
+	if ((rfmt == 1) && !css_general_characteristics.fcs)
+		return -EINVAL;
+	if ((rfmt == 2) && !css_general_characteristics.cib)
+		return -EINVAL;
+	scpd_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!scpd_area)
+		return -ENOMEM;
+
+	scpd_area->request.length = 0x0010;
+	scpd_area->request.code = 0x0002;
+
+	scpd_area->cssid = chpid.cssid;
+	scpd_area->first_chpid = chpid.id;
+	scpd_area->last_chpid = chpid.id;
+	scpd_area->m = m;
+	scpd_area->c = c;
+	scpd_area->fmt = fmt;
+	scpd_area->rfmt = rfmt;
+
+	ccode = chsc(scpd_area);
+	if (ccode > 0) {
+		ret = (ccode == 3) ? -ENODEV : -EBUSY;
+		goto out;
+	}
+
+	ret = chsc_error_from_response(scpd_area->response.code);
+	if (ret == 0)
+		/* Success. */
+		memcpy(resp, &scpd_area->response, scpd_area->response.length);
+	else
+		CIO_CRW_EVENT(2, "chsc: scpd failed (rc=%04x)\n",
+			      scpd_area->response.code);
+out:
+	free_page((unsigned long)scpd_area);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(chsc_determine_channel_path_desc);
+
+int chsc_determine_base_channel_path_desc(struct chp_id chpid,
+					  struct channel_path_desc *desc)
+{
+	struct chsc_response_struct *chsc_resp;
+	int ret;
+
+	chsc_resp = kzalloc(sizeof(*chsc_resp), GFP_KERNEL);
+	if (!chsc_resp)
+		return -ENOMEM;
+	ret = chsc_determine_channel_path_desc(chpid, 0, 0, 0, 0, chsc_resp);
+	if (ret)
+		goto out_free;
+	memcpy(desc, &chsc_resp->data, chsc_resp->length);
+out_free:
+	kfree(chsc_resp);
+	return ret;
+}
+
+static void
+chsc_initialize_cmg_chars(struct channel_path *chp, u8 cmcv,
+			  struct cmg_chars *chars)
+{
+	switch (chp->cmg) {
+	case 2:
+	case 3:
+		chp->cmg_chars = kmalloc(sizeof(struct cmg_chars),
+					 GFP_KERNEL);
+		if (chp->cmg_chars) {
+			int i, mask;
+			struct cmg_chars *cmg_chars;
+
+			cmg_chars = chp->cmg_chars;
+			for (i = 0; i < NR_MEASUREMENT_CHARS; i++) {
+				mask = 0x80 >> (i + 3);
+				if (cmcv & mask)
+					cmg_chars->values[i] = chars->values[i];
+				else
+					cmg_chars->values[i] = 0;
+			}
+		}
+		break;
+	default:
+		/* No cmg-dependent data. */
+		break;
+	}
+}
+
+int chsc_get_channel_measurement_chars(struct channel_path *chp)
+{
+	int ccode, ret;
+
+	struct {
+		struct chsc_header request;
+		u32 : 24;
+		u32 first_chpid : 8;
+		u32 : 24;
+		u32 last_chpid : 8;
+		u32 zeroes1;
+		struct chsc_header response;
+		u32 zeroes2;
+		u32 not_valid : 1;
+		u32 shared : 1;
+		u32 : 22;
+		u32 chpid : 8;
+		u32 cmcv : 5;
+		u32 : 11;
+		u32 cmgq : 8;
+		u32 cmg : 8;
+		u32 zeroes3;
+		u32 data[NR_MEASUREMENT_CHARS];
+	} __attribute__ ((packed)) *scmc_area;
+
+	scmc_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!scmc_area)
+		return -ENOMEM;
+
+	scmc_area->request.length = 0x0010;
+	scmc_area->request.code = 0x0022;
+
+	scmc_area->first_chpid = chp->chpid.id;
+	scmc_area->last_chpid = chp->chpid.id;
+
+	ccode = chsc(scmc_area);
+	if (ccode > 0) {
+		ret = (ccode == 3) ? -ENODEV : -EBUSY;
+		goto out;
+	}
+
+	ret = chsc_error_from_response(scmc_area->response.code);
+	if (ret == 0) {
+		/* Success. */
+		if (!scmc_area->not_valid) {
+			chp->cmg = scmc_area->cmg;
+			chp->shared = scmc_area->shared;
+			chsc_initialize_cmg_chars(chp, scmc_area->cmcv,
+						  (struct cmg_chars *)
+						  &scmc_area->data);
+		} else {
+			chp->cmg = -1;
+			chp->shared = -1;
+		}
+	} else {
+		CIO_CRW_EVENT(2, "chsc: scmc failed (rc=%04x)\n",
+			      scmc_area->response.code);
+	}
+out:
+	free_page((unsigned long)scmc_area);
+	return ret;
+}
+
+int __init chsc_alloc_sei_area(void)
+{
+	int ret;
+
+	sei_page = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sei_page) {
+		CIO_MSG_EVENT(0, "Can't allocate page for processing of "
+			      "chsc machine checks!\n");
+		return -ENOMEM;
+	}
+	ret = crw_register_handler(CRW_RSC_CSS, chsc_process_crw);
+	if (ret)
+		kfree(sei_page);
+	return ret;
+}
+
+void __init chsc_free_sei_area(void)
+{
+	crw_unregister_handler(CRW_RSC_CSS);
+	kfree(sei_page);
+}
+
+int __init
+chsc_enable_facility(int operation_code)
+{
+	int ret;
+	struct {
+		struct chsc_header request;
+		u8 reserved1:4;
+		u8 format:4;
+		u8 reserved2;
+		u16 operation_code;
+		u32 reserved3;
+		u32 reserved4;
+		u32 operation_data_area[252];
+		struct chsc_header response;
+		u32 reserved5:4;
+		u32 format2:4;
+		u32 reserved6:24;
+	} __attribute__ ((packed)) *sda_area;
+
+	sda_area = (void *)get_zeroed_page(GFP_KERNEL|GFP_DMA);
+	if (!sda_area)
+		return -ENOMEM;
+	sda_area->request.length = 0x0400;
+	sda_area->request.code = 0x0031;
+	sda_area->operation_code = operation_code;
+
+	ret = chsc(sda_area);
+	if (ret > 0) {
+		ret = (ret == 3) ? -ENODEV : -EBUSY;
+		goto out;
+	}
+
+	switch (sda_area->response.code) {
+	case 0x0101:
+		ret = -EOPNOTSUPP;
+		break;
+	default:
+		ret = chsc_error_from_response(sda_area->response.code);
+	}
+	if (ret != 0)
+		CIO_CRW_EVENT(2, "chsc: sda (oc=%x) failed (rc=%04x)\n",
+			      operation_code, sda_area->response.code);
+ out:
+	free_page((unsigned long)sda_area);
+	return ret;
+}
+
+struct css_general_char css_general_characteristics;
+struct css_chsc_char css_chsc_characteristics;
+
+int __init
+chsc_determine_css_characteristics(void)
+{
+	int result;
+	struct {
+		struct chsc_header request;
+		u32 reserved1;
+		u32 reserved2;
+		u32 reserved3;
+		struct chsc_header response;
+		u32 reserved4;
+		u32 general_char[510];
+		u32 chsc_char[518];
+	} __attribute__ ((packed)) *scsc_area;
+
+	scsc_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!scsc_area)
+		return -ENOMEM;
+
+	scsc_area->request.length = 0x0010;
+	scsc_area->request.code = 0x0010;
+
+	result = chsc(scsc_area);
+	if (result) {
+		result = (result == 3) ? -ENODEV : -EBUSY;
+		goto exit;
+	}
+
+	result = chsc_error_from_response(scsc_area->response.code);
+	if (result == 0) {
+		memcpy(&css_general_characteristics, scsc_area->general_char,
+		       sizeof(css_general_characteristics));
+		memcpy(&css_chsc_characteristics, scsc_area->chsc_char,
+		       sizeof(css_chsc_characteristics));
+	} else
+		CIO_CRW_EVENT(2, "chsc: scsc failed (rc=%04x)\n",
+			      scsc_area->response.code);
+exit:
+	free_page ((unsigned long) scsc_area);
+	return result;
+}
+
+EXPORT_SYMBOL_GPL(css_general_characteristics);
+EXPORT_SYMBOL_GPL(css_chsc_characteristics);
+
+int chsc_sstpc(void *page, unsigned int op, u16 ctrl)
+{
+	struct {
+		struct chsc_header request;
+		unsigned int rsvd0;
+		unsigned int op : 8;
+		unsigned int rsvd1 : 8;
+		unsigned int ctrl : 16;
+		unsigned int rsvd2[5];
+		struct chsc_header response;
+		unsigned int rsvd3[7];
+	} __attribute__ ((packed)) *rr;
+	int rc;
+
+	memset(page, 0, PAGE_SIZE);
+	rr = page;
+	rr->request.length = 0x0020;
+	rr->request.code = 0x0033;
+	rr->op = op;
+	rr->ctrl = ctrl;
+	rc = chsc(rr);
+	if (rc)
+		return -EIO;
+	rc = (rr->response.code == 0x0001) ? 0 : -EIO;
+	return rc;
+}
+
+int chsc_sstpi(void *page, void *result, size_t size)
+{
+	struct {
+		struct chsc_header request;
+		unsigned int rsvd0[3];
+		struct chsc_header response;
+		char data[size];
+	} __attribute__ ((packed)) *rr;
+	int rc;
+
+	memset(page, 0, PAGE_SIZE);
+	rr = page;
+	rr->request.length = 0x0010;
+	rr->request.code = 0x0038;
+	rc = chsc(rr);
+	if (rc)
+		return -EIO;
+	memcpy(result, &rr->data, size);
+	return (rr->response.code == 0x0001) ? 0 : -EIO;
+}
+

kernel/drivers/s390/cio/chsc.h

@@ -0,0 +1,83 @@
+#ifndef S390_CHSC_H
+#define S390_CHSC_H
+
+#include <linux/types.h>
+#include <linux/device.h>
+#include <asm/chpid.h>
+#include <asm/chsc.h>
+#include <asm/schid.h>
+
+#define CHSC_SDA_OC_MSS   0x2
+
+struct chsc_header {
+	u16 length;
+	u16 code;
+} __attribute__ ((packed));
+
+#define NR_MEASUREMENT_CHARS 5
+struct cmg_chars {
+	u32 values[NR_MEASUREMENT_CHARS];
+} __attribute__ ((packed));
+
+#define NR_MEASUREMENT_ENTRIES 8
+struct cmg_entry {
+	u32 values[NR_MEASUREMENT_ENTRIES];
+} __attribute__ ((packed));
+
+struct channel_path_desc {
+	u8 flags;
+	u8 lsn;
+	u8 desc;
+	u8 chpid;
+	u8 swla;
+	u8 zeroes;
+	u8 chla;
+	u8 chpp;
+} __attribute__ ((packed));
+
+struct channel_path;
+
+struct css_chsc_char {
+	u64 res;
+	u64 : 20;
+	u32 secm : 1; /* bit 84 */
+	u32 : 1;
+	u32 scmc : 1; /* bit 86 */
+	u32 : 20;
+	u32 scssc : 1;  /* bit 107 */
+	u32 scsscf : 1; /* bit 108 */
+	u32 : 19;
+}__attribute__((packed));
+
+extern struct css_chsc_char css_chsc_characteristics;
+
+struct chsc_ssd_info {
+	u8 path_mask;
+	u8 fla_valid_mask;
+	struct chp_id chpid[8];
+	u16 fla[8];
+};
+extern int chsc_get_ssd_info(struct subchannel_id schid,
+			     struct chsc_ssd_info *ssd);
+extern int chsc_determine_css_characteristics(void);
+extern int chsc_alloc_sei_area(void);
+extern void chsc_free_sei_area(void);
+
+extern int chsc_enable_facility(int);
+struct channel_subsystem;
+extern int chsc_secm(struct channel_subsystem *, int);
+int __chsc_do_secm(struct channel_subsystem *css, int enable, void *page);
+
+int chsc_chp_vary(struct chp_id chpid, int on);
+int chsc_determine_channel_path_desc(struct chp_id chpid, int fmt, int rfmt,
+				     int c, int m,
+				     struct chsc_response_struct *resp);
+int chsc_determine_base_channel_path_desc(struct chp_id chpid,
+					  struct channel_path_desc *desc);
+void chsc_chp_online(struct chp_id chpid);
+void chsc_chp_offline(struct chp_id chpid);
+int chsc_get_channel_measurement_chars(struct channel_path *chp);
+
+int chsc_error_from_response(int response);
+
+#endif

kernel/drivers/s390/cio/chsc_sch.c

@@ -0,0 +1,850 @@
+/*
+ * Driver for s390 chsc subchannels
+ *
+ * Copyright IBM Corp. 2008, 2009
+ *
+ * Author(s): Cornelia Huck <cornelia.huck@de.ibm.com>
+ *
+ */
+
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/uaccess.h>
+#include <linux/miscdevice.h>
+
+#include <asm/cio.h>
+#include <asm/chsc.h>
+#include <asm/isc.h>
+
+#include "cio.h"
+#include "cio_debug.h"
+#include "css.h"
+#include "chsc_sch.h"
+#include "ioasm.h"
+
+static debug_info_t *chsc_debug_msg_id;
+static debug_info_t *chsc_debug_log_id;
+
+#define CHSC_MSG(imp, args...) do {					\
+		debug_sprintf_event(chsc_debug_msg_id, imp , ##args);	\
+	} while (0)
+
+#define CHSC_LOG(imp, txt) do {					\
+		debug_text_event(chsc_debug_log_id, imp , txt);	\
+	} while (0)
+
+static void CHSC_LOG_HEX(int level, void *data, int length)
+{
+	while (length > 0) {
+		debug_event(chsc_debug_log_id, level, data, length);
+		length -= chsc_debug_log_id->buf_size;
+		data += chsc_debug_log_id->buf_size;
+	}
+}
+
+MODULE_AUTHOR("IBM Corporation");
+MODULE_DESCRIPTION("driver for s390 chsc subchannels");
+MODULE_LICENSE("GPL");
+
+static void chsc_subchannel_irq(struct subchannel *sch)
+{
+	struct chsc_private *private = sch->private;
+	struct chsc_request *request = private->request;
+	struct irb *irb = (struct irb *)__LC_IRB;
+
+	CHSC_LOG(4, "irb");
+	CHSC_LOG_HEX(4, irb, sizeof(*irb));
+	/* Copy irb to provided request and set done. */
+	if (!request) {
+		CHSC_MSG(0, "Interrupt on sch 0.%x.%04x with no request\n",
+			 sch->schid.ssid, sch->schid.sch_no);
+		return;
+	}
+	private->request = NULL;
+	memcpy(&request->irb, irb, sizeof(*irb));
+	cio_update_schib(sch);
+	complete(&request->completion);
+	put_device(&sch->dev);
+}
+
+static int chsc_subchannel_probe(struct subchannel *sch)
+{
+	struct chsc_private *private;
+	int ret;
+
+	CHSC_MSG(6, "Detected chsc subchannel 0.%x.%04x\n",
+		 sch->schid.ssid, sch->schid.sch_no);
+	sch->isc = CHSC_SCH_ISC;
+	private = kzalloc(sizeof(*private), GFP_KERNEL);
+	if (!private)
+		return -ENOMEM;
+	ret = cio_enable_subchannel(sch, (u32)(unsigned long)sch);
+	if (ret) {
+		CHSC_MSG(0, "Failed to enable 0.%x.%04x: %d\n",
+			 sch->schid.ssid, sch->schid.sch_no, ret);
+		kfree(private);
+	} else {
+		sch->private = private;
+		if (dev_get_uevent_suppress(&sch->dev)) {
+			dev_set_uevent_suppress(&sch->dev, 0);
+			kobject_uevent(&sch->dev.kobj, KOBJ_ADD);
+		}
+	}
+	return ret;
+}
+
+static int chsc_subchannel_remove(struct subchannel *sch)
+{
+	struct chsc_private *private;
+
+	cio_disable_subchannel(sch);
+	private = sch->private;
+	sch->private = NULL;
+	if (private->request) {
+		complete(&private->request->completion);
+		put_device(&sch->dev);
+	}
+	kfree(private);
+	return 0;
+}
+
+static void chsc_subchannel_shutdown(struct subchannel *sch)
+{
+	cio_disable_subchannel(sch);
+}
+
+static int chsc_subchannel_prepare(struct subchannel *sch)
+{
+	int cc;
+	struct schib schib;
+	/*
+	 * Don't allow suspend while the subchannel is not idle
+	 * since we don't have a way to clear the subchannel and
+	 * cannot disable it with a request running.
+	 */
+	cc = stsch(sch->schid, &schib);
+	if (!cc && scsw_stctl(&schib.scsw))
+		return -EAGAIN;
+	return 0;
+}
+
+static int chsc_subchannel_freeze(struct subchannel *sch)
+{
+	return cio_disable_subchannel(sch);
+}
+
+static int chsc_subchannel_restore(struct subchannel *sch)
+{
+	return cio_enable_subchannel(sch, (u32)(unsigned long)sch);
+}
+
+static struct css_device_id chsc_subchannel_ids[] = {
+	{ .match_flags = 0x1, .type =SUBCHANNEL_TYPE_CHSC, },
+	{ /* end of list */ },
+};
+MODULE_DEVICE_TABLE(css, chsc_subchannel_ids);
+
+static struct css_driver chsc_subchannel_driver = {
+	.owner = THIS_MODULE,
+	.subchannel_type = chsc_subchannel_ids,
+	.irq = chsc_subchannel_irq,
+	.probe = chsc_subchannel_probe,
+	.remove = chsc_subchannel_remove,
+	.shutdown = chsc_subchannel_shutdown,
+	.prepare = chsc_subchannel_prepare,
+	.freeze = chsc_subchannel_freeze,
+	.thaw = chsc_subchannel_restore,
+	.restore = chsc_subchannel_restore,
+	.name = "chsc_subchannel",
+};
+
+static int __init chsc_init_dbfs(void)
+{
+	chsc_debug_msg_id = debug_register("chsc_msg", 16, 1,
+					   16 * sizeof(long));
+	if (!chsc_debug_msg_id)
+		goto out;
+	debug_register_view(chsc_debug_msg_id, &debug_sprintf_view);
+	debug_set_level(chsc_debug_msg_id, 2);
+	chsc_debug_log_id = debug_register("chsc_log", 16, 1, 16);
+	if (!chsc_debug_log_id)
+		goto out;
+	debug_register_view(chsc_debug_log_id, &debug_hex_ascii_view);
+	debug_set_level(chsc_debug_log_id, 2);
+	return 0;
+out:
+	if (chsc_debug_msg_id)
+		debug_unregister(chsc_debug_msg_id);
+	return -ENOMEM;
+}
+
+static void chsc_remove_dbfs(void)
+{
+	debug_unregister(chsc_debug_log_id);
+	debug_unregister(chsc_debug_msg_id);
+}
+
+static int __init chsc_init_sch_driver(void)
+{
+	return css_driver_register(&chsc_subchannel_driver);
+}
+
+static void chsc_cleanup_sch_driver(void)
+{
+	css_driver_unregister(&chsc_subchannel_driver);
+}
+
+static DEFINE_SPINLOCK(chsc_lock);
+
+static int chsc_subchannel_match_next_free(struct device *dev, void *data)
+{
+	struct subchannel *sch = to_subchannel(dev);
+
+	return sch->schib.pmcw.ena && !scsw_fctl(&sch->schib.scsw);
+}
+
+static struct subchannel *chsc_get_next_subchannel(struct subchannel *sch)
+{
+	struct device *dev;
+
+	dev = driver_find_device(&chsc_subchannel_driver.drv,
+				 sch ? &sch->dev : NULL, NULL,
+				 chsc_subchannel_match_next_free);
+	return dev ? to_subchannel(dev) : NULL;
+}
+
+/**
+ * chsc_async() - try to start a chsc request asynchronously
+ * @chsc_area: request to be started
+ * @request: request structure to associate
+ *
+ * Tries to start a chsc request on one of the existing chsc subchannels.
+ * Returns:
+ *  %0 if the request was performed synchronously
+ *  %-EINPROGRESS if the request was successfully started
+ *  %-EBUSY if all chsc subchannels are busy
+ *  %-ENODEV if no chsc subchannels are available
+ * Context:
+ *  interrupts disabled, chsc_lock held
+ */
+static int chsc_async(struct chsc_async_area *chsc_area,
+		      struct chsc_request *request)
+{
+	int cc;
+	struct chsc_private *private;
+	struct subchannel *sch = NULL;
+	int ret = -ENODEV;
+	char dbf[10];
+
+	chsc_area->header.key = PAGE_DEFAULT_KEY;
+	while ((sch = chsc_get_next_subchannel(sch))) {
+		spin_lock(sch->lock);
+		private = sch->private;
+		if (private->request) {
+			spin_unlock(sch->lock);
+			ret = -EBUSY;
+			continue;
+		}
+		chsc_area->header.sid = sch->schid;
+		CHSC_LOG(2, "schid");
+		CHSC_LOG_HEX(2, &sch->schid, sizeof(sch->schid));
+		cc = chsc(chsc_area);
+		sprintf(dbf, "cc:%d", cc);
+		CHSC_LOG(2, dbf);
+		switch (cc) {
+		case 0:
+			ret = 0;
+			break;
+		case 1:
+			sch->schib.scsw.cmd.fctl |= SCSW_FCTL_START_FUNC;
+			ret = -EINPROGRESS;
+			private->request = request;
+			break;
+		case 2:
+			ret = -EBUSY;
+			break;
+		default:
+			ret = -ENODEV;
+		}
+		spin_unlock(sch->lock);
+		CHSC_MSG(2, "chsc on 0.%x.%04x returned cc=%d\n",
+			 sch->schid.ssid, sch->schid.sch_no, cc);
+		if (ret == -EINPROGRESS)
+			return -EINPROGRESS;
+		put_device(&sch->dev);
+		if (ret == 0)
+			return 0;
+	}
+	return ret;
+}
+
+static void chsc_log_command(struct chsc_async_area *chsc_area)
+{
+	char dbf[10];
+
+	sprintf(dbf, "CHSC:%x", chsc_area->header.code);
+	CHSC_LOG(0, dbf);
+	CHSC_LOG_HEX(0, chsc_area, 32);
+}
+
+static int chsc_examine_irb(struct chsc_request *request)
+{
+	int backed_up;
+
+	if (!(scsw_stctl(&request->irb.scsw) & SCSW_STCTL_STATUS_PEND))
+		return -EIO;
+	backed_up = scsw_cstat(&request->irb.scsw) & SCHN_STAT_CHAIN_CHECK;
+	request->irb.scsw.cmd.cstat &= ~SCHN_STAT_CHAIN_CHECK;
+	if (scsw_cstat(&request->irb.scsw) == 0)
+		return 0;
+	if (!backed_up)
+		return 0;
+	if (scsw_cstat(&request->irb.scsw) & SCHN_STAT_PROG_CHECK)
+		return -EIO;
+	if (scsw_cstat(&request->irb.scsw) & SCHN_STAT_PROT_CHECK)
+		return -EPERM;
+	if (scsw_cstat(&request->irb.scsw) & SCHN_STAT_CHN_DATA_CHK)
+		return -EAGAIN;
+	if (scsw_cstat(&request->irb.scsw) & SCHN_STAT_CHN_CTRL_CHK)
+		return -EAGAIN;
+	return -EIO;
+}
+
+static int chsc_ioctl_start(void __user *user_area)
+{
+	struct chsc_request *request;
+	struct chsc_async_area *chsc_area;
+	int ret;
+	char dbf[10];
+
+	if (!css_general_characteristics.dynio)
+		/* It makes no sense to try. */
+		return -EOPNOTSUPP;
+	chsc_area = (void *)get_zeroed_page(GFP_DMA | GFP_KERNEL);
+	if (!chsc_area)
+		return -ENOMEM;
+	request = kzalloc(sizeof(*request), GFP_KERNEL);
+	if (!request) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+	init_completion(&request->completion);
+	if (copy_from_user(chsc_area, user_area, PAGE_SIZE)) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+	chsc_log_command(chsc_area);
+	spin_lock_irq(&chsc_lock);
+	ret = chsc_async(chsc_area, request);
+	spin_unlock_irq(&chsc_lock);
+	if (ret == -EINPROGRESS) {
+		wait_for_completion(&request->completion);
+		ret = chsc_examine_irb(request);
+	}
+	/* copy area back to user */
+	if (!ret)
+		if (copy_to_user(user_area, chsc_area, PAGE_SIZE))
+			ret = -EFAULT;
+out_free:
+	sprintf(dbf, "ret:%d", ret);
+	CHSC_LOG(0, dbf);
+	kfree(request);
+	free_page((unsigned long)chsc_area);
+	return ret;
+}
+
+static int chsc_ioctl_info_channel_path(void __user *user_cd)
+{
+	struct chsc_chp_cd *cd;
+	int ret, ccode;
+	struct {
+		struct chsc_header request;
+		u32 : 2;
+		u32 m : 1;
+		u32 : 1;
+		u32 fmt1 : 4;
+		u32 cssid : 8;
+		u32 : 8;
+		u32 first_chpid : 8;
+		u32 : 24;
+		u32 last_chpid : 8;
+		u32 : 32;
+		struct chsc_header response;
+		u8 data[PAGE_SIZE - 20];
+	} __attribute__ ((packed)) *scpcd_area;
+
+	scpcd_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!scpcd_area)
+		return -ENOMEM;
+	cd = kzalloc(sizeof(*cd), GFP_KERNEL);
+	if (!cd) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+	if (copy_from_user(cd, user_cd, sizeof(*cd))) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+	scpcd_area->request.length = 0x0010;
+	scpcd_area->request.code = 0x0028;
+	scpcd_area->m = cd->m;
+	scpcd_area->fmt1 = cd->fmt;
+	scpcd_area->cssid = cd->chpid.cssid;
+	scpcd_area->first_chpid = cd->chpid.id;
+	scpcd_area->last_chpid = cd->chpid.id;
+
+	ccode = chsc(scpcd_area);
+	if (ccode != 0) {
+		ret = -EIO;
+		goto out_free;
+	}
+	if (scpcd_area->response.code != 0x0001) {
+		ret = -EIO;
+		CHSC_MSG(0, "scpcd: response code=%x\n",
+			 scpcd_area->response.code);
+		goto out_free;
+	}
+	memcpy(&cd->cpcb, &scpcd_area->response, scpcd_area->response.length);
+	if (copy_to_user(user_cd, cd, sizeof(*cd)))
+		ret = -EFAULT;
+	else
+		ret = 0;
+out_free:
+	kfree(cd);
+	free_page((unsigned long)scpcd_area);
+	return ret;
+}
+
+static int chsc_ioctl_info_cu(void __user *user_cd)
+{
+	struct chsc_cu_cd *cd;
+	int ret, ccode;
+	struct {
+		struct chsc_header request;
+		u32 : 2;
+		u32 m : 1;
+		u32 : 1;
+		u32 fmt1 : 4;
+		u32 cssid : 8;
+		u32 : 8;
+		u32 first_cun : 8;
+		u32 : 24;
+		u32 last_cun : 8;
+		u32 : 32;
+		struct chsc_header response;
+		u8 data[PAGE_SIZE - 20];
+	} __attribute__ ((packed)) *scucd_area;
+
+	scucd_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!scucd_area)
+		return -ENOMEM;
+	cd = kzalloc(sizeof(*cd), GFP_KERNEL);
+	if (!cd) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+	if (copy_from_user(cd, user_cd, sizeof(*cd))) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+	scucd_area->request.length = 0x0010;
+	scucd_area->request.code = 0x0028;
+	scucd_area->m = cd->m;
+	scucd_area->fmt1 = cd->fmt;
+	scucd_area->cssid = cd->cssid;
+	scucd_area->first_cun = cd->cun;
+	scucd_area->last_cun = cd->cun;
+
+	ccode = chsc(scucd_area);
+	if (ccode != 0) {
+		ret = -EIO;
+		goto out_free;
+	}
+	if (scucd_area->response.code != 0x0001) {
+		ret = -EIO;
+		CHSC_MSG(0, "scucd: response code=%x\n",
+			 scucd_area->response.code);
+		goto out_free;
+	}
+	memcpy(&cd->cucb, &scucd_area->response, scucd_area->response.length);
+	if (copy_to_user(user_cd, cd, sizeof(*cd)))
+		ret = -EFAULT;
+	else
+		ret = 0;
+out_free:
+	kfree(cd);
+	free_page((unsigned long)scucd_area);
+	return ret;
+}
+
+static int chsc_ioctl_info_sch_cu(void __user *user_cud)
+{
+	struct chsc_sch_cud *cud;
+	int ret, ccode;
+	struct {
+		struct chsc_header request;
+		u32 : 2;
+		u32 m : 1;
+		u32 : 5;
+		u32 fmt1 : 4;
+		u32 : 2;
+		u32 ssid : 2;
+		u32 first_sch : 16;
+		u32 : 8;
+		u32 cssid : 8;
+		u32 last_sch : 16;
+		u32 : 32;
+		struct chsc_header response;
+		u8 data[PAGE_SIZE - 20];
+	} __attribute__ ((packed)) *sscud_area;
+
+	sscud_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sscud_area)
+		return -ENOMEM;
+	cud = kzalloc(sizeof(*cud), GFP_KERNEL);
+	if (!cud) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+	if (copy_from_user(cud, user_cud, sizeof(*cud))) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+	sscud_area->request.length = 0x0010;
+	sscud_area->request.code = 0x0006;
+	sscud_area->m = cud->schid.m;
+	sscud_area->fmt1 = cud->fmt;
+	sscud_area->ssid = cud->schid.ssid;
+	sscud_area->first_sch = cud->schid.sch_no;
+	sscud_area->cssid = cud->schid.cssid;
+	sscud_area->last_sch = cud->schid.sch_no;
+
+	ccode = chsc(sscud_area);
+	if (ccode != 0) {
+		ret = -EIO;
+		goto out_free;
+	}
+	if (sscud_area->response.code != 0x0001) {
+		ret = -EIO;
+		CHSC_MSG(0, "sscud: response code=%x\n",
+			 sscud_area->response.code);
+		goto out_free;
+	}
+	memcpy(&cud->scub, &sscud_area->response, sscud_area->response.length);
+	if (copy_to_user(user_cud, cud, sizeof(*cud)))
+		ret = -EFAULT;
+	else
+		ret = 0;
+out_free:
+	kfree(cud);
+	free_page((unsigned long)sscud_area);
+	return ret;
+}
+
+static int chsc_ioctl_conf_info(void __user *user_ci)
+{
+	struct chsc_conf_info *ci;
+	int ret, ccode;
+	struct {
+		struct chsc_header request;
+		u32 : 2;
+		u32 m : 1;
+		u32 : 1;
+		u32 fmt1 : 4;
+		u32 cssid : 8;
+		u32 : 6;
+		u32 ssid : 2;
+		u32 : 8;
+		u64 : 64;
+		struct chsc_header response;
+		u8 data[PAGE_SIZE - 20];
+	} __attribute__ ((packed)) *sci_area;
+
+	sci_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sci_area)
+		return -ENOMEM;
+	ci = kzalloc(sizeof(*ci), GFP_KERNEL);
+	if (!ci) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+	if (copy_from_user(ci, user_ci, sizeof(*ci))) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+	sci_area->request.length = 0x0010;
+	sci_area->request.code = 0x0012;
+	sci_area->m = ci->id.m;
+	sci_area->fmt1 = ci->fmt;
+	sci_area->cssid = ci->id.cssid;
+	sci_area->ssid = ci->id.ssid;
+
+	ccode = chsc(sci_area);
+	if (ccode != 0) {
+		ret = -EIO;
+		goto out_free;
+	}
+	if (sci_area->response.code != 0x0001) {
+		ret = -EIO;
+		CHSC_MSG(0, "sci: response code=%x\n",
+			 sci_area->response.code);
+		goto out_free;
+	}
+	memcpy(&ci->scid, &sci_area->response, sci_area->response.length);
+	if (copy_to_user(user_ci, ci, sizeof(*ci)))
+		ret = -EFAULT;
+	else
+		ret = 0;
+out_free:
+	kfree(ci);
+	free_page((unsigned long)sci_area);
+	return ret;
+}
+
+static int chsc_ioctl_conf_comp_list(void __user *user_ccl)
+{
+	struct chsc_comp_list *ccl;
+	int ret, ccode;
+	struct {
+		struct chsc_header request;
+		u32 ctype : 8;
+		u32 : 4;
+		u32 fmt : 4;
+		u32 : 16;
+		u64 : 64;
+		u32 list_parm[2];
+		u64 : 64;
+		struct chsc_header response;
+		u8 data[PAGE_SIZE - 36];
+	} __attribute__ ((packed)) *sccl_area;
+	struct {
+		u32 m : 1;
+		u32 : 31;
+		u32 cssid : 8;
+		u32 : 16;
+		u32 chpid : 8;
+	} __attribute__ ((packed)) *chpid_parm;
+	struct {
+		u32 f_cssid : 8;
+		u32 l_cssid : 8;
+		u32 : 16;
+		u32 res;
+	} __attribute__ ((packed)) *cssids_parm;
+
+	sccl_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sccl_area)
+		return -ENOMEM;
+	ccl = kzalloc(sizeof(*ccl), GFP_KERNEL);
+	if (!ccl) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+	if (copy_from_user(ccl, user_ccl, sizeof(*ccl))) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+	sccl_area->request.length = 0x0020;
+	sccl_area->request.code = 0x0030;
+	sccl_area->fmt = ccl->req.fmt;
+	sccl_area->ctype = ccl->req.ctype;
+	switch (sccl_area->ctype) {
+	case CCL_CU_ON_CHP:
+	case CCL_IOP_CHP:
+		chpid_parm = (void *)&sccl_area->list_parm;
+		chpid_parm->m = ccl->req.chpid.m;
+		chpid_parm->cssid = ccl->req.chpid.chp.cssid;
+		chpid_parm->chpid = ccl->req.chpid.chp.id;
+		break;
+	case CCL_CSS_IMG:
+	case CCL_CSS_IMG_CONF_CHAR:
+		cssids_parm = (void *)&sccl_area->list_parm;
+		cssids_parm->f_cssid = ccl->req.cssids.f_cssid;
+		cssids_parm->l_cssid = ccl->req.cssids.l_cssid;
+		break;
+	}
+	ccode = chsc(sccl_area);
+	if (ccode != 0) {
+		ret = -EIO;
+		goto out_free;
+	}
+	if (sccl_area->response.code != 0x0001) {
+		ret = -EIO;
+		CHSC_MSG(0, "sccl: response code=%x\n",
+			 sccl_area->response.code);
+		goto out_free;
+	}
+	memcpy(&ccl->sccl, &sccl_area->response, sccl_area->response.length);
+	if (copy_to_user(user_ccl, ccl, sizeof(*ccl)))
+		ret = -EFAULT;
+	else
+		ret = 0;
+out_free:
+	kfree(ccl);
+	free_page((unsigned long)sccl_area);
+	return ret;
+}
+
+static int chsc_ioctl_chpd(void __user *user_chpd)
+{
+	struct chsc_cpd_info *chpd;
+	int ret;
+
+	chpd = kzalloc(sizeof(*chpd), GFP_KERNEL);
+	if (!chpd)
+		return -ENOMEM;
+	if (copy_from_user(chpd, user_chpd, sizeof(*chpd))) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+	ret = chsc_determine_channel_path_desc(chpd->chpid, chpd->fmt,
+					       chpd->rfmt, chpd->c, chpd->m,
+					       &chpd->chpdb);
+	if (ret)
+		goto out_free;
+	if (copy_to_user(user_chpd, chpd, sizeof(*chpd)))
+		ret = -EFAULT;
+out_free:
+	kfree(chpd);
+	return ret;
+}
+
+static int chsc_ioctl_dcal(void __user *user_dcal)
+{
+	struct chsc_dcal *dcal;
+	int ret, ccode;
+	struct {
+		struct chsc_header request;
+		u32 atype : 8;
+		u32 : 4;
+		u32 fmt : 4;
+		u32 : 16;
+		u32 res0[2];
+		u32 list_parm[2];
+		u32 res1[2];
+		struct chsc_header response;
+		u8 data[PAGE_SIZE - 36];
+	} __attribute__ ((packed)) *sdcal_area;
+
+	sdcal_area = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
+	if (!sdcal_area)
+		return -ENOMEM;
+	dcal = kzalloc(sizeof(*dcal), GFP_KERNEL);
+	if (!dcal) {
+		ret = -ENOMEM;
+		goto out_free;
+	}
+	if (copy_from_user(dcal, user_dcal, sizeof(*dcal))) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+	sdcal_area->request.length = 0x0020;
+	sdcal_area->request.code = 0x0034;
+	sdcal_area->atype = dcal->req.atype;
+	sdcal_area->fmt = dcal->req.fmt;
+	memcpy(&sdcal_area->list_parm, &dcal->req.list_parm,
+	       sizeof(sdcal_area->list_parm));
+
+	ccode = chsc(sdcal_area);
+	if (ccode != 0) {
+		ret = -EIO;
+		goto out_free;
+	}
+	if (sdcal_area->response.code != 0x0001) {
+		ret = -EIO;
+		CHSC_MSG(0, "sdcal: response code=%x\n",
+			 sdcal_area->response.code);
+		goto out_free;
+	}
+	memcpy(&dcal->sdcal, &sdcal_area->response,
+	       sdcal_area->response.length);
+	if (copy_to_user(user_dcal, dcal, sizeof(*dcal)))
+		ret = -EFAULT;
+	else
+		ret = 0;
+out_free:
+	kfree(dcal);
+	free_page((unsigned long)sdcal_area);
+	return ret;
+}
+
+static long chsc_ioctl(struct file *filp, unsigned int cmd,
+		       unsigned long arg)
+{
+	CHSC_MSG(2, "chsc_ioctl called, cmd=%x\n", cmd);
+	switch (cmd) {
+	case CHSC_START:
+		return chsc_ioctl_start((void __user *)arg);
+	case CHSC_INFO_CHANNEL_PATH:
+		return chsc_ioctl_info_channel_path((void __user *)arg);
+	case CHSC_INFO_CU:
+		return chsc_ioctl_info_cu((void __user *)arg);
+	case CHSC_INFO_SCH_CU:
+		return chsc_ioctl_info_sch_cu((void __user *)arg);
+	case CHSC_INFO_CI:
+		return chsc_ioctl_conf_info((void __user *)arg);
+	case CHSC_INFO_CCL:
+		return chsc_ioctl_conf_comp_list((void __user *)arg);
+	case CHSC_INFO_CPD:
+		return chsc_ioctl_chpd((void __user *)arg);
+	case CHSC_INFO_DCAL:
+		return chsc_ioctl_dcal((void __user *)arg);
+	default: /* unknown ioctl number */
+		return -ENOIOCTLCMD;
+	}
+}
+
+static const struct file_operations chsc_fops = {
+	.owner = THIS_MODULE,
+	.unlocked_ioctl = chsc_ioctl,
+	.compat_ioctl = chsc_ioctl,
+};
+
+static struct miscdevice chsc_misc_device = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "chsc",
+	.fops = &chsc_fops,
+};
+
+static int __init chsc_misc_init(void)
+{
+	return misc_register(&chsc_misc_device);
+}
+
+static void chsc_misc_cleanup(void)
+{
+	misc_deregister(&chsc_misc_device);
+}
+
+static int __init chsc_sch_init(void)
+{
+	int ret;
+
+	ret = chsc_init_dbfs();
+	if (ret)
+		return ret;
+	isc_register(CHSC_SCH_ISC);
+	ret = chsc_init_sch_driver();
+	if (ret)
+		goto out_dbf;
+	ret = chsc_misc_init();
+	if (ret)
+		goto out_driver;
+	return ret;
+out_driver:
+	chsc_cleanup_sch_driver();
+out_dbf:
+	isc_unregister(CHSC_SCH_ISC);
+	chsc_remove_dbfs();
+	return ret;
+}
+
+static void __exit chsc_sch_exit(void)
+{
+	chsc_misc_cleanup();
+	chsc_cleanup_sch_driver();
+	isc_unregister(CHSC_SCH_ISC);
+	chsc_remove_dbfs();
+}
+
+module_init(chsc_sch_init);
+module_exit(chsc_sch_exit);

kernel/drivers/s390/cio/chsc_sch.h

@@ -0,0 +1,13 @@
+#ifndef _CHSC_SCH_H
+#define _CHSC_SCH_H
+
+struct chsc_request {
+	struct completion completion;
+	struct irb irb;
+};
+
+struct chsc_private {
+	struct chsc_request *request;
+};
+
+#endif

kernel/drivers/s390/cio/cio.h

@@ -0,0 +1,143 @@
+#ifndef S390_CIO_H
+#define S390_CIO_H
+
+#include <linux/mutex.h>
+#include <linux/device.h>
+#include <linux/mod_devicetable.h>
+#include <asm/chpid.h>
+#include <asm/cio.h>
+#include <asm/fcx.h>
+#include <asm/schid.h>
+#include "chsc.h"
+
+/*
+ * path management control word
+ */
+struct pmcw {
+	u32 intparm;		/* interruption parameter */
+	u32 qf	 : 1;		/* qdio facility */
+	u32 w	 : 1;
+	u32 isc  : 3;		/* interruption sublass */
+	u32 res5 : 3;		/* reserved zeros */
+	u32 ena  : 1;		/* enabled */
+	u32 lm	 : 2;		/* limit mode */
+	u32 mme  : 2;		/* measurement-mode enable */
+	u32 mp	 : 1;		/* multipath mode */
+	u32 tf	 : 1;		/* timing facility */
+	u32 dnv  : 1;		/* device number valid */
+	u32 dev  : 16;		/* device number */
+	u8  lpm;		/* logical path mask */
+	u8  pnom;		/* path not operational mask */
+	u8  lpum;		/* last path used mask */
+	u8  pim;		/* path installed mask */
+	u16 mbi;		/* measurement-block index */
+	u8  pom;		/* path operational mask */
+	u8  pam;		/* path available mask */
+	u8  chpid[8];		/* CHPID 0-7 (if available) */
+	u32 unused1 : 8;	/* reserved zeros */
+	u32 st	    : 3;	/* subchannel type */
+	u32 unused2 : 18;	/* reserved zeros */
+	u32 mbfc    : 1;	/* measurement block format control */
+	u32 xmwme   : 1;	/* extended measurement word mode enable */
+	u32 csense  : 1;	/* concurrent sense; can be enabled ...*/
+				/*  ... per MSCH, however, if facility */
+				/*  ... is not installed, this results */
+				/*  ... in an operand exception.       */
+} __attribute__ ((packed));
+
+/* Target SCHIB configuration. */
+struct schib_config {
+	u64 mba;
+	u32 intparm;
+	u16 mbi;
+	u32 isc:3;
+	u32 ena:1;
+	u32 mme:2;
+	u32 mp:1;
+	u32 csense:1;
+	u32 mbfc:1;
+} __attribute__ ((packed));
+
+/*
+ * subchannel information block
+ */
+struct schib {
+	struct pmcw pmcw;	 /* path management control word */
+	union scsw scsw;	 /* subchannel status word */
+	__u64 mba;               /* measurement block address */
+	__u8 mda[4];		 /* model dependent area */
+} __attribute__ ((packed,aligned(4)));
+
+/* subchannel data structure used by I/O subroutines */
+struct subchannel {
+	struct subchannel_id schid;
+	spinlock_t *lock;	/* subchannel lock */
+	struct mutex reg_mutex;
+	enum {
+		SUBCHANNEL_TYPE_IO = 0,
+		SUBCHANNEL_TYPE_CHSC = 1,
+		SUBCHANNEL_TYPE_MSG = 2,
+		SUBCHANNEL_TYPE_ADM = 3,
+	} st;			/* subchannel type */
+
+	struct {
+		unsigned int suspend:1; /* allow suspend */
+		unsigned int prefetch:1;/* deny prefetch */
+		unsigned int inter:1;   /* suppress intermediate interrupts */
+	} __attribute__ ((packed)) options;
+
+	__u8 vpm;		/* verified path mask */
+	__u8 lpm;		/* logical path mask */
+	__u8 opm;               /* operational path mask */
+	struct schib schib;	/* subchannel information block */
+	int isc; /* desired interruption subclass */
+	struct chsc_ssd_info ssd_info;	/* subchannel description */
+	struct device dev;	/* entry in device tree */
+	struct css_driver *driver;
+	void *private; /* private per subchannel type data */
+	struct work_struct work;
+	struct schib_config config;
+} __attribute__ ((aligned(8)));
+
+#define IO_INTERRUPT_TYPE	   0 /* I/O interrupt type */
+
+#define to_subchannel(n) container_of(n, struct subchannel, dev)
+
+extern int cio_validate_subchannel (struct subchannel *, struct subchannel_id);
+extern int cio_enable_subchannel(struct subchannel *, u32);
+extern int cio_disable_subchannel (struct subchannel *);
+extern int cio_cancel (struct subchannel *);
+extern int cio_clear (struct subchannel *);
+extern int cio_resume (struct subchannel *);
+extern int cio_halt (struct subchannel *);
+extern int cio_start (struct subchannel *, struct ccw1 *, __u8);
+extern int cio_start_key (struct subchannel *, struct ccw1 *, __u8, __u8);
+extern int cio_cancel (struct subchannel *);
+extern int cio_set_options (struct subchannel *, int);
+extern int cio_get_options (struct subchannel *);
+extern int cio_update_schib(struct subchannel *sch);
+extern int cio_commit_config(struct subchannel *sch);
+
+int cio_tm_start_key(struct subchannel *sch, struct tcw *tcw, u8 lpm, u8 key);
+int cio_tm_intrg(struct subchannel *sch);
+
+int cio_create_sch_lock(struct subchannel *);
+void do_adapter_IO(u8 isc);
+void do_IRQ(struct pt_regs *);
+
+/* Use with care. */
+#ifdef CONFIG_CCW_CONSOLE
+extern struct subchannel *cio_probe_console(void);
+extern void cio_release_console(void);
+extern int cio_is_console(struct subchannel_id);
+extern struct subchannel *cio_get_console_subchannel(void);
+extern spinlock_t * cio_get_console_lock(void);
+extern void *cio_get_console_priv(void);
+#else
+#define cio_is_console(schid) 0
+#define cio_get_console_subchannel() NULL
+#define cio_get_console_lock() NULL
+#define cio_get_console_priv() NULL
+#endif
+
+#endif

kernel/drivers/s390/cio/cio_debug.h

@@ -0,0 +1,34 @@
+#ifndef CIO_DEBUG_H
+#define CIO_DEBUG_H
+
+#include <asm/debug.h>
+
+/* for use of debug feature */
+extern debug_info_t *cio_debug_msg_id;
+extern debug_info_t *cio_debug_trace_id;
+extern debug_info_t *cio_debug_crw_id;
+
+#define CIO_TRACE_EVENT(imp, txt) do {				\
+		debug_text_event(cio_debug_trace_id, imp, txt); \
+	} while (0)
+
+#define CIO_MSG_EVENT(imp, args...) do {				\
+		debug_sprintf_event(cio_debug_msg_id, imp , ##args);	\
+	} while (0)
+
+#define CIO_CRW_EVENT(imp, args...) do {				\
+		debug_sprintf_event(cio_debug_crw_id, imp , ##args);	\
+	} while (0)
+
+static inline void CIO_HEX_EVENT(int level, void *data, int length)
+{
+	if (unlikely(!cio_debug_trace_id))
+		return;
+	while (length > 0) {
+		debug_event(cio_debug_trace_id, level, data, length);
+		length -= cio_debug_trace_id->buf_size;
+		data += cio_debug_trace_id->buf_size;
+	}
+}
+
+#endif

kernel/drivers/s390/cio/crw.c

@@ -0,0 +1,159 @@
+/*
+ *   Channel report handling code
+ *
+ *    Copyright IBM Corp. 2000,2009
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>,
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>,
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>,
+ */
+
+#include <linux/semaphore.h>
+#include <linux/mutex.h>
+#include <linux/kthread.h>
+#include <linux/init.h>
+#include <asm/crw.h>
+
+static struct semaphore crw_semaphore;
+static DEFINE_MUTEX(crw_handler_mutex);
+static crw_handler_t crw_handlers[NR_RSCS];
+
+/**
+ * crw_register_handler() - register a channel report word handler
+ * @rsc: reporting source code to handle
+ * @handler: handler to be registered
+ *
+ * Returns %0 on success and a negative error value otherwise.
+ */
+int crw_register_handler(int rsc, crw_handler_t handler)
+{
+	int rc = 0;
+
+	if ((rsc < 0) || (rsc >= NR_RSCS))
+		return -EINVAL;
+	mutex_lock(&crw_handler_mutex);
+	if (crw_handlers[rsc])
+		rc = -EBUSY;
+	else
+		crw_handlers[rsc] = handler;
+	mutex_unlock(&crw_handler_mutex);
+	return rc;
+}
+
+/**
+ * crw_unregister_handler() - unregister a channel report word handler
+ * @rsc: reporting source code to handle
+ */
+void crw_unregister_handler(int rsc)
+{
+	if ((rsc < 0) || (rsc >= NR_RSCS))
+		return;
+	mutex_lock(&crw_handler_mutex);
+	crw_handlers[rsc] = NULL;
+	mutex_unlock(&crw_handler_mutex);
+}
+
+/*
+ * Retrieve CRWs and call function to handle event.
+ */
+static int crw_collect_info(void *unused)
+{
+	struct crw crw[2];
+	int ccode;
+	unsigned int chain;
+	int ignore;
+
+repeat:
+	ignore = down_interruptible(&crw_semaphore);
+	chain = 0;
+	while (1) {
+		crw_handler_t handler;
+
+		if (unlikely(chain > 1)) {
+			struct crw tmp_crw;
+
+			printk(KERN_WARNING"%s: Code does not support more "
+			       "than two chained crws; please report to "
+			       "linux390@de.ibm.com!\n", __func__);
+			ccode = stcrw(&tmp_crw);
+			printk(KERN_WARNING"%s: crw reports slct=%d, oflw=%d, "
+			       "chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
+			       __func__, tmp_crw.slct, tmp_crw.oflw,
+			       tmp_crw.chn, tmp_crw.rsc, tmp_crw.anc,
+			       tmp_crw.erc, tmp_crw.rsid);
+			printk(KERN_WARNING"%s: This was crw number %x in the "
+			       "chain\n", __func__, chain);
+			if (ccode != 0)
+				break;
+			chain = tmp_crw.chn ? chain + 1 : 0;
+			continue;
+		}
+		ccode = stcrw(&crw[chain]);
+		if (ccode != 0)
+			break;
+		printk(KERN_DEBUG "crw_info : CRW reports slct=%d, oflw=%d, "
+		       "chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
+		       crw[chain].slct, crw[chain].oflw, crw[chain].chn,
+		       crw[chain].rsc, crw[chain].anc, crw[chain].erc,
+		       crw[chain].rsid);
+		/* Check for overflows. */
+		if (crw[chain].oflw) {
+			int i;
+
+			pr_debug("%s: crw overflow detected!\n", __func__);
+			mutex_lock(&crw_handler_mutex);
+			for (i = 0; i < NR_RSCS; i++) {
+				if (crw_handlers[i])
+					crw_handlers[i](NULL, NULL, 1);
+			}
+			mutex_unlock(&crw_handler_mutex);
+			chain = 0;
+			continue;
+		}
+		if (crw[0].chn && !chain) {
+			chain++;
+			continue;
+		}
+		mutex_lock(&crw_handler_mutex);
+		handler = crw_handlers[crw[chain].rsc];
+		if (handler)
+			handler(&crw[0], chain ? &crw[1] : NULL, 0);
+		mutex_unlock(&crw_handler_mutex);
+		/* chain is always 0 or 1 here. */
+		chain = crw[chain].chn ? chain + 1 : 0;
+	}
+	goto repeat;
+	return 0;
+}
+
+void crw_handle_channel_report(void)
+{
+	up(&crw_semaphore);
+}
+
+/*
+ * Separate initcall needed for semaphore initialization since
+ * crw_handle_channel_report might be called before crw_machine_check_init.
+ */
+static int __init crw_init_semaphore(void)
+{
+	init_MUTEX_LOCKED(&crw_semaphore);
+	return 0;
+}
+pure_initcall(crw_init_semaphore);
+
+/*
+ * Machine checks for the channel subsystem must be enabled
+ * after the channel subsystem is initialized
+ */
+static int __init crw_machine_check_init(void)
+{
+	struct task_struct *task;
+
+	task = kthread_run(crw_collect_info, NULL, "kmcheck");
+	if (IS_ERR(task))
+		return PTR_ERR(task);
+	ctl_set_bit(14, 28);	/* enable channel report MCH */
+	return 0;
+}
+device_initcall(crw_machine_check_init);

kernel/drivers/s390/cio/css.h

@@ -0,0 +1,154 @@
+#ifndef _CSS_H
+#define _CSS_H
+
+#include <linux/mutex.h>
+#include <linux/wait.h>
+#include <linux/workqueue.h>
+#include <linux/device.h>
+#include <linux/types.h>
+
+#include <asm/cio.h>
+#include <asm/chpid.h>
+#include <asm/schid.h>
+
+/*
+ * path grouping stuff
+ */
+#define SPID_FUNC_SINGLE_PATH	   0x00
+#define SPID_FUNC_MULTI_PATH	   0x80
+#define SPID_FUNC_ESTABLISH	   0x00
+#define SPID_FUNC_RESIGN	   0x40
+#define SPID_FUNC_DISBAND	   0x20
+
+#define SNID_STATE1_RESET	   0
+#define SNID_STATE1_UNGROUPED	   2
+#define SNID_STATE1_GROUPED	   3
+
+#define SNID_STATE2_NOT_RESVD	   0
+#define SNID_STATE2_RESVD_ELSE	   2
+#define SNID_STATE2_RESVD_SELF	   3
+
+#define SNID_STATE3_MULTI_PATH	   1
+#define SNID_STATE3_SINGLE_PATH	   0
+
+struct path_state {
+	__u8  state1 : 2;	/* path state value 1 */
+	__u8  state2 : 2;	/* path state value 2 */
+	__u8  state3 : 1;	/* path state value 3 */
+	__u8  resvd  : 3;	/* reserved */
+} __attribute__ ((packed));
+
+struct extended_cssid {
+	u8 version;
+	u8 cssid;
+} __attribute__ ((packed));
+
+struct pgid {
+	union {
+		__u8 fc;   	/* SPID function code */
+		struct path_state ps;	/* SNID path state */
+	} __attribute__ ((packed)) inf;
+	union {
+		__u32 cpu_addr	: 16;	/* CPU address */
+		struct extended_cssid ext_cssid;
+	} __attribute__ ((packed)) pgid_high;
+	__u32 cpu_id	: 24;	/* CPU identification */
+	__u32 cpu_model : 16;	/* CPU model */
+	__u32 tod_high;		/* high word TOD clock */
+} __attribute__ ((packed));
+
+struct subchannel;
+struct chp_link;
+/**
+ * struct css_driver - device driver for subchannels
+ * @owner: owning module
+ * @subchannel_type: subchannel type supported by this driver
+ * @drv: embedded device driver structure
+ * @irq: called on interrupts
+ * @chp_event: called for events affecting a channel path
+ * @sch_event: called for events affecting the subchannel
+ * @probe: function called on probe
+ * @remove: function called on remove
+ * @shutdown: called at device shutdown
+ * @prepare: prepare for pm state transition
+ * @complete: undo work done in @prepare
+ * @freeze: callback for freezing during hibernation snapshotting
+ * @thaw: undo work done in @freeze
+ * @restore: callback for restoring after hibernation
+ * @settle: wait for asynchronous work to finish
+ * @name: name of the device driver
+ */
+struct css_driver {
+	struct module *owner;
+	struct css_device_id *subchannel_type;
+	struct device_driver drv;
+	void (*irq)(struct subchannel *);
+	int (*chp_event)(struct subchannel *, struct chp_link *, int);
+	int (*sch_event)(struct subchannel *, int);
+	int (*probe)(struct subchannel *);
+	int (*remove)(struct subchannel *);
+	void (*shutdown)(struct subchannel *);
+	int (*prepare) (struct subchannel *);
+	void (*complete) (struct subchannel *);
+	int (*freeze)(struct subchannel *);
+	int (*thaw) (struct subchannel *);
+	int (*restore)(struct subchannel *);
+	void (*settle)(void);
+	const char *name;
+};
+
+#define to_cssdriver(n) container_of(n, struct css_driver, drv)
+
+/*
+ * all css_drivers have the css_bus_type
+ */
+extern struct bus_type css_bus_type;
+
+extern int css_driver_register(struct css_driver *);
+extern void css_driver_unregister(struct css_driver *);
+
+extern void css_sch_device_unregister(struct subchannel *);
+extern int css_probe_device(struct subchannel_id);
+extern struct subchannel *get_subchannel_by_schid(struct subchannel_id);
+extern int css_init_done;
+extern int max_ssid;
+int for_each_subchannel_staged(int (*fn_known)(struct subchannel *, void *),
+			       int (*fn_unknown)(struct subchannel_id,
+			       void *), void *data);
+extern int for_each_subchannel(int(*fn)(struct subchannel_id, void *), void *);
+extern void css_reiterate_subchannels(void);
+void css_update_ssd_info(struct subchannel *sch);
+
+#define __MAX_SUBCHANNEL 65535
+#define __MAX_SSID 3
+
+struct channel_subsystem {
+	u8 cssid;
+	int valid;
+	struct channel_path *chps[__MAX_CHPID + 1];
+	struct device device;
+	struct pgid global_pgid;
+	struct mutex mutex;
+	/* channel measurement related */
+	int cm_enabled;
+	void *cub_addr1;
+	void *cub_addr2;
+	/* for orphaned ccw devices */
+	struct subchannel *pseudo_subchannel;
+};
+#define to_css(dev) container_of(dev, struct channel_subsystem, device)
+
+extern struct bus_type css_bus_type;
+extern struct channel_subsystem *channel_subsystems[];
+
+/* Helper functions to build lists for the slow path. */
+void css_schedule_eval(struct subchannel_id schid);
+void css_schedule_eval_all(void);
+
+int sch_is_pseudo_sch(struct subchannel *);
+struct schib;
+int css_sch_is_valid(struct schib *);
+
+extern struct workqueue_struct *slow_path_wq;
+void css_wait_for_slow_path(void);
+#endif

kernel/drivers/s390/cio/device.h

@@ -0,0 +1,142 @@
+#ifndef S390_DEVICE_H
+#define S390_DEVICE_H
+
+#include <asm/ccwdev.h>
+#include <asm/atomic.h>
+#include <linux/wait.h>
+
+#include "io_sch.h"
+
+/*
+ * states of the device statemachine
+ */
+enum dev_state {
+	DEV_STATE_NOT_OPER,
+	DEV_STATE_SENSE_PGID,
+	DEV_STATE_SENSE_ID,
+	DEV_STATE_OFFLINE,
+	DEV_STATE_VERIFY,
+	DEV_STATE_ONLINE,
+	DEV_STATE_W4SENSE,
+	DEV_STATE_DISBAND_PGID,
+	DEV_STATE_BOXED,
+	/* states to wait for i/o completion before doing something */
+	DEV_STATE_CLEAR_VERIFY,
+	DEV_STATE_TIMEOUT_KILL,
+	DEV_STATE_QUIESCE,
+	/* special states for devices gone not operational */
+	DEV_STATE_DISCONNECTED,
+	DEV_STATE_DISCONNECTED_SENSE_ID,
+	DEV_STATE_CMFCHANGE,
+	DEV_STATE_CMFUPDATE,
+	/* last element! */
+	NR_DEV_STATES
+};
+
+/*
+ * asynchronous events of the device statemachine
+ */
+enum dev_event {
+	DEV_EVENT_NOTOPER,
+	DEV_EVENT_INTERRUPT,
+	DEV_EVENT_TIMEOUT,
+	DEV_EVENT_VERIFY,
+	/* last element! */
+	NR_DEV_EVENTS
+};
+
+struct ccw_device;
+
+/*
+ * action called through jumptable
+ */
+typedef void (fsm_func_t)(struct ccw_device *, enum dev_event);
+extern fsm_func_t *dev_jumptable[NR_DEV_STATES][NR_DEV_EVENTS];
+
+static inline void
+dev_fsm_event(struct ccw_device *cdev, enum dev_event dev_event)
+{
+	dev_jumptable[cdev->private->state][dev_event](cdev, dev_event);
+}
+
+/*
+ * Delivers 1 if the device state is final.
+ */
+static inline int
+dev_fsm_final_state(struct ccw_device *cdev)
+{
+	return (cdev->private->state == DEV_STATE_NOT_OPER ||
+		cdev->private->state == DEV_STATE_OFFLINE ||
+		cdev->private->state == DEV_STATE_ONLINE ||
+		cdev->private->state == DEV_STATE_BOXED);
+}
+
+extern struct workqueue_struct *ccw_device_work;
+extern wait_queue_head_t ccw_device_init_wq;
+extern atomic_t ccw_device_init_count;
+int __init io_subchannel_init(void);
+
+void io_subchannel_recog_done(struct ccw_device *cdev);
+void io_subchannel_init_config(struct subchannel *sch);
+
+int ccw_device_cancel_halt_clear(struct ccw_device *);
+
+void ccw_device_do_unbind_bind(struct work_struct *);
+void ccw_device_move_to_orphanage(struct work_struct *);
+int ccw_device_is_orphan(struct ccw_device *);
+
+int ccw_device_recognition(struct ccw_device *);
+int ccw_device_online(struct ccw_device *);
+int ccw_device_offline(struct ccw_device *);
+void ccw_device_update_sense_data(struct ccw_device *);
+int ccw_device_test_sense_data(struct ccw_device *);
+void ccw_device_schedule_sch_unregister(struct ccw_device *);
+int ccw_purge_blacklisted(void);
+
+/* Function prototypes for device status and basic sense stuff. */
+void ccw_device_accumulate_irb(struct ccw_device *, struct irb *);
+void ccw_device_accumulate_basic_sense(struct ccw_device *, struct irb *);
+int ccw_device_accumulate_and_sense(struct ccw_device *, struct irb *);
+int ccw_device_do_sense(struct ccw_device *, struct irb *);
+
+/* Function prototypes for sense id stuff. */
+void ccw_device_sense_id_start(struct ccw_device *);
+void ccw_device_sense_id_irq(struct ccw_device *, enum dev_event);
+void ccw_device_sense_id_done(struct ccw_device *, int);
+
+/* Function prototypes for path grouping stuff. */
+void ccw_device_sense_pgid_start(struct ccw_device *);
+void ccw_device_sense_pgid_irq(struct ccw_device *, enum dev_event);
+void ccw_device_sense_pgid_done(struct ccw_device *, int);
+
+void ccw_device_verify_start(struct ccw_device *);
+void ccw_device_verify_irq(struct ccw_device *, enum dev_event);
+void ccw_device_verify_done(struct ccw_device *, int);
+
+void ccw_device_disband_start(struct ccw_device *);
+void ccw_device_disband_irq(struct ccw_device *, enum dev_event);
+void ccw_device_disband_done(struct ccw_device *, int);
+
+int ccw_device_call_handler(struct ccw_device *);
+
+int ccw_device_stlck(struct ccw_device *);
+
+/* Helper function for machine check handling. */
+void ccw_device_trigger_reprobe(struct ccw_device *);
+void ccw_device_kill_io(struct ccw_device *);
+int ccw_device_notify(struct ccw_device *, int);
+void ccw_device_set_disconnected(struct ccw_device *cdev);
+void ccw_device_set_notoper(struct ccw_device *cdev);
+
+/* qdio needs this. */
+void ccw_device_set_timeout(struct ccw_device *, int);
+extern struct subchannel_id ccw_device_get_subchannel_id(struct ccw_device *);
+extern struct bus_type ccw_bus_type;
+
+/* Channel measurement facility related */
+void retry_set_schib(struct ccw_device *cdev);
+void cmf_retry_copy_block(struct ccw_device *);
+int cmf_reenable(struct ccw_device *);
+int ccw_set_cmf(struct ccw_device *cdev, int enable);
+extern struct device_attribute dev_attr_cmb_enable;
+#endif

kernel/drivers/s390/cio/device_id.c

@@ -0,0 +1,323 @@
+/*
+ * drivers/s390/cio/device_id.c
+ *
+ *    Copyright (C) 2002 IBM Deutschland Entwicklung GmbH,
+ *			 IBM Corporation
+ *    Author(s): Cornelia Huck (cornelia.huck@de.ibm.com)
+ *		 Martin Schwidefsky (schwidefsky@de.ibm.com)
+ *
+ * Sense ID functions.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+
+#include <asm/ccwdev.h>
+#include <asm/delay.h>
+#include <asm/cio.h>
+#include <asm/lowcore.h>
+#include <asm/diag.h>
+
+#include "cio.h"
+#include "cio_debug.h"
+#include "css.h"
+#include "device.h"
+#include "ioasm.h"
+#include "io_sch.h"
+
+/**
+ * vm_vdev_to_cu_type - Convert vm virtual device into control unit type
+ *			for certain devices.
+ * @class: virtual device class
+ * @type: virtual device type
+ *
+ * Returns control unit type if a match was made or %0xffff otherwise.
+ */
+static int vm_vdev_to_cu_type(int class, int type)
+{
+	static struct {
+		int class, type, cu_type;
+	} vm_devices[] = {
+		{ 0x08, 0x01, 0x3480 },
+		{ 0x08, 0x02, 0x3430 },
+		{ 0x08, 0x10, 0x3420 },
+		{ 0x08, 0x42, 0x3424 },
+		{ 0x08, 0x44, 0x9348 },
+		{ 0x08, 0x81, 0x3490 },
+		{ 0x08, 0x82, 0x3422 },
+		{ 0x10, 0x41, 0x1403 },
+		{ 0x10, 0x42, 0x3211 },
+		{ 0x10, 0x43, 0x3203 },
+		{ 0x10, 0x45, 0x3800 },
+		{ 0x10, 0x47, 0x3262 },
+		{ 0x10, 0x48, 0x3820 },
+		{ 0x10, 0x49, 0x3800 },
+		{ 0x10, 0x4a, 0x4245 },
+		{ 0x10, 0x4b, 0x4248 },
+		{ 0x10, 0x4d, 0x3800 },
+		{ 0x10, 0x4e, 0x3820 },
+		{ 0x10, 0x4f, 0x3820 },
+		{ 0x10, 0x82, 0x2540 },
+		{ 0x10, 0x84, 0x3525 },
+		{ 0x20, 0x81, 0x2501 },
+		{ 0x20, 0x82, 0x2540 },
+		{ 0x20, 0x84, 0x3505 },
+		{ 0x40, 0x01, 0x3278 },
+		{ 0x40, 0x04, 0x3277 },
+		{ 0x40, 0x80, 0x2250 },
+		{ 0x40, 0xc0, 0x5080 },
+		{ 0x80, 0x00, 0x3215 },
+	};
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(vm_devices); i++)
+		if (class == vm_devices[i].class && type == vm_devices[i].type)
+			return vm_devices[i].cu_type;
+
+	return 0xffff;
+}
+
+/**
+ * diag_get_dev_info - retrieve device information via DIAG X'210'
+ * @devno: device number
+ * @ps: pointer to sense ID data area
+ *
+ * Returns zero on success, non-zero otherwise.
+ */
+static int diag_get_dev_info(u16 devno, struct senseid *ps)
+{
+	struct diag210 diag_data;
+	int ccode;
+
+	CIO_TRACE_EVENT (4, "VMvdinf");
+
+	diag_data = (struct diag210) {
+		.vrdcdvno = devno,
+		.vrdclen = sizeof (diag_data),
+	};
+
+	ccode = diag210 (&diag_data);
+	if ((ccode == 0) || (ccode == 2)) {
+		ps->reserved = 0xff;
+
+		/* Special case for osa devices. */
+		if (diag_data.vrdcvcla == 0x02 && diag_data.vrdcvtyp == 0x20) {
+			ps->cu_type = 0x3088;
+			ps->cu_model = 0x60;
+			return 0;
+		}
+		ps->cu_type = vm_vdev_to_cu_type(diag_data.vrdcvcla,
+						diag_data.vrdcvtyp);
+		if (ps->cu_type != 0xffff)
+			return 0;
+	}
+
+	CIO_MSG_EVENT(0, "DIAG X'210' for device %04X returned (cc = %d):"
+		      "vdev class : %02X, vdev type : %04X \n ...  "
+		      "rdev class : %02X, rdev type : %04X, "
+		      "rdev model: %02X\n",
+		      devno, ccode,
+		      diag_data.vrdcvcla, diag_data.vrdcvtyp,
+		      diag_data.vrdcrccl, diag_data.vrdccrty,
+		      diag_data.vrdccrmd);
+
+	return -ENODEV;
+}
+
+/*
+ * Start Sense ID helper function.
+ * Try to obtain the 'control unit'/'device type' information
+ *  associated with the subchannel.
+ */
+static int
+__ccw_device_sense_id_start(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	struct ccw1 *ccw;
+	int ret;
+
+	sch = to_subchannel(cdev->dev.parent);
+	/* Setup sense channel program. */
+	ccw = cdev->private->iccws;
+	ccw->cmd_code = CCW_CMD_SENSE_ID;
+	ccw->cda = (__u32) __pa (&cdev->private->senseid);
+	ccw->count = sizeof (struct senseid);
+	ccw->flags = CCW_FLAG_SLI;
+
+	/* Reset device status. */
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+
+	/* Try on every path. */
+	ret = -ENODEV;
+	while (cdev->private->imask != 0) {
+		cdev->private->senseid.cu_type = 0xFFFF;
+		if ((sch->opm & cdev->private->imask) != 0 &&
+		    cdev->private->iretry > 0) {
+			cdev->private->iretry--;
+			/* Reset internal retry indication. */
+			cdev->private->flags.intretry = 0;
+			ret = cio_start (sch, cdev->private->iccws,
+					 cdev->private->imask);
+			/* ret is 0, -EBUSY, -EACCES or -ENODEV */
+			if (ret != -EACCES)
+				return ret;
+		}
+		cdev->private->imask >>= 1;
+		cdev->private->iretry = 5;
+	}
+	return ret;
+}
+
+void
+ccw_device_sense_id_start(struct ccw_device *cdev)
+{
+	int ret;
+
+	memset (&cdev->private->senseid, 0, sizeof (struct senseid));
+	cdev->private->imask = 0x80;
+	cdev->private->iretry = 5;
+	ret = __ccw_device_sense_id_start(cdev);
+	if (ret && ret != -EBUSY)
+		ccw_device_sense_id_done(cdev, ret);
+}
+
+/*
+ * Called from interrupt context to check if a valid answer
+ * to Sense ID was received.
+ */
+static int
+ccw_device_check_sense_id(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	struct irb *irb;
+
+	sch = to_subchannel(cdev->dev.parent);
+	irb = &cdev->private->irb;
+
+	/* Check the error cases. */
+	if (irb->scsw.cmd.fctl & (SCSW_FCTL_HALT_FUNC | SCSW_FCTL_CLEAR_FUNC)) {
+		/* Retry Sense ID if requested. */
+		if (cdev->private->flags.intretry) {
+			cdev->private->flags.intretry = 0;
+			return -EAGAIN;
+		}
+		return -ETIME;
+	}
+	if (irb->esw.esw0.erw.cons && (irb->ecw[0] & SNS0_CMD_REJECT)) {
+		/*
+		 * if the device doesn't support the SenseID
+		 *  command further retries wouldn't help ...
+		 * NB: We don't check here for intervention required like we
+		 *     did before, because tape devices with no tape inserted
+		 *     may present this status *in conjunction with* the
+		 *     sense id information. So, for intervention required,
+		 *     we use the "whack it until it talks" strategy...
+		 */
+		CIO_MSG_EVENT(0, "SenseID : device %04x on Subchannel "
+			      "0.%x.%04x reports cmd reject\n",
+			      cdev->private->dev_id.devno, sch->schid.ssid,
+			      sch->schid.sch_no);
+		return -EOPNOTSUPP;
+	}
+	if (irb->esw.esw0.erw.cons) {
+		CIO_MSG_EVENT(2, "SenseID : UC on dev 0.%x.%04x, "
+			      "lpum %02X, cnt %02d, sns :"
+			      " %02X%02X%02X%02X %02X%02X%02X%02X ...\n",
+			      cdev->private->dev_id.ssid,
+			      cdev->private->dev_id.devno,
+			      irb->esw.esw0.sublog.lpum,
+			      irb->esw.esw0.erw.scnt,
+			      irb->ecw[0], irb->ecw[1],
+			      irb->ecw[2], irb->ecw[3],
+			      irb->ecw[4], irb->ecw[5],
+			      irb->ecw[6], irb->ecw[7]);
+		return -EAGAIN;
+	}
+	if (irb->scsw.cmd.cc == 3) {
+		u8 lpm;
+
+		lpm = to_io_private(sch)->orb.cmd.lpm;
+		if ((lpm & sch->schib.pmcw.pim & sch->schib.pmcw.pam) != 0)
+			CIO_MSG_EVENT(4, "SenseID : path %02X for device %04x "
+				      "on subchannel 0.%x.%04x is "
+				      "'not operational'\n", lpm,
+				      cdev->private->dev_id.devno,
+				      sch->schid.ssid, sch->schid.sch_no);
+		return -EACCES;
+	}
+
+	/* Did we get a proper answer ? */
+	if (irb->scsw.cmd.cc == 0 && cdev->private->senseid.cu_type != 0xFFFF &&
+	    cdev->private->senseid.reserved == 0xFF) {
+		if (irb->scsw.cmd.count < sizeof(struct senseid) - 8)
+			cdev->private->flags.esid = 1;
+		return 0; /* Success */
+	}
+
+	/* Hmm, whatever happened, try again. */
+	CIO_MSG_EVENT(2, "SenseID : start_IO() for device %04x on "
+		      "subchannel 0.%x.%04x returns status %02X%02X\n",
+		      cdev->private->dev_id.devno, sch->schid.ssid,
+		      sch->schid.sch_no,
+		      irb->scsw.cmd.dstat, irb->scsw.cmd.cstat);
+	return -EAGAIN;
+}
+
+/*
+ * Got interrupt for Sense ID.
+ */
+void
+ccw_device_sense_id_irq(struct ccw_device *cdev, enum dev_event dev_event)
+{
+	struct subchannel *sch;
+	struct irb *irb;
+	int ret;
+
+	sch = to_subchannel(cdev->dev.parent);
+	irb = (struct irb *) __LC_IRB;
+	/* Retry sense id, if needed. */
+	if (irb->scsw.cmd.stctl ==
+	    (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)) {
+		if ((irb->scsw.cmd.cc == 1) || !irb->scsw.cmd.actl) {
+			ret = __ccw_device_sense_id_start(cdev);
+			if (ret && ret != -EBUSY)
+				ccw_device_sense_id_done(cdev, ret);
+		}
+		return;
+	}
+	if (ccw_device_accumulate_and_sense(cdev, irb) != 0)
+		return;
+	ret = ccw_device_check_sense_id(cdev);
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+	switch (ret) {
+	/* 0, -ETIME, -EOPNOTSUPP, -EAGAIN or -EACCES */
+	case 0:			/* Sense id succeeded. */
+	case -ETIME:		/* Sense id stopped by timeout. */
+		ccw_device_sense_id_done(cdev, ret);
+		break;
+	case -EACCES:		/* channel is not operational. */
+		sch->lpm &= ~cdev->private->imask;
+		cdev->private->imask >>= 1;
+		cdev->private->iretry = 5;
+		/* fall through. */
+	case -EAGAIN:		/* try again. */
+		ret = __ccw_device_sense_id_start(cdev);
+		if (ret == 0 || ret == -EBUSY)
+			break;
+		/* fall through. */
+	default:		/* Sense ID failed. Try asking VM. */
+		if (MACHINE_IS_VM)
+			ret = diag_get_dev_info(cdev->private->dev_id.devno,
+						&cdev->private->senseid);
+		else
+			/*
+			 * If we can't couldn't identify the device type we
+			 *  consider the device "not operational".
+			 */
+			ret = -ENODEV;
+
+		ccw_device_sense_id_done(cdev, ret);
+		break;
+	}
+}

kernel/drivers/s390/cio/device_ops.c

@@ -0,0 +1,716 @@
+/*
+ * Copyright IBM Corp. 2002, 2009
+ *
+ * Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
+ *	      Cornelia Huck (cornelia.huck@de.ibm.com)
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/list.h>
+#include <linux/device.h>
+#include <linux/delay.h>
+
+#include <asm/ccwdev.h>
+#include <asm/idals.h>
+#include <asm/chpid.h>
+#include <asm/fcx.h>
+
+#include "cio.h"
+#include "cio_debug.h"
+#include "css.h"
+#include "chsc.h"
+#include "device.h"
+#include "chp.h"
+
+/**
+ * ccw_device_set_options_mask() - set some options and unset the rest
+ * @cdev: device for which the options are to be set
+ * @flags: options to be set
+ *
+ * All flags specified in @flags are set, all flags not specified in @flags
+ * are cleared.
+ * Returns:
+ *   %0 on success, -%EINVAL on an invalid flag combination.
+ */
+int ccw_device_set_options_mask(struct ccw_device *cdev, unsigned long flags)
+{
+       /*
+	* The flag usage is mutal exclusive ...
+	*/
+	if ((flags & CCWDEV_EARLY_NOTIFICATION) &&
+	    (flags & CCWDEV_REPORT_ALL))
+		return -EINVAL;
+	cdev->private->options.fast = (flags & CCWDEV_EARLY_NOTIFICATION) != 0;
+	cdev->private->options.repall = (flags & CCWDEV_REPORT_ALL) != 0;
+	cdev->private->options.pgroup = (flags & CCWDEV_DO_PATHGROUP) != 0;
+	cdev->private->options.force = (flags & CCWDEV_ALLOW_FORCE) != 0;
+	return 0;
+}
+
+/**
+ * ccw_device_set_options() - set some options
+ * @cdev: device for which the options are to be set
+ * @flags: options to be set
+ *
+ * All flags specified in @flags are set, the remainder is left untouched.
+ * Returns:
+ *   %0 on success, -%EINVAL if an invalid flag combination would ensue.
+ */
+int ccw_device_set_options(struct ccw_device *cdev, unsigned long flags)
+{
+       /*
+	* The flag usage is mutal exclusive ...
+	*/
+	if (((flags & CCWDEV_EARLY_NOTIFICATION) &&
+	    (flags & CCWDEV_REPORT_ALL)) ||
+	    ((flags & CCWDEV_EARLY_NOTIFICATION) &&
+	     cdev->private->options.repall) ||
+	    ((flags & CCWDEV_REPORT_ALL) &&
+	     cdev->private->options.fast))
+		return -EINVAL;
+	cdev->private->options.fast |= (flags & CCWDEV_EARLY_NOTIFICATION) != 0;
+	cdev->private->options.repall |= (flags & CCWDEV_REPORT_ALL) != 0;
+	cdev->private->options.pgroup |= (flags & CCWDEV_DO_PATHGROUP) != 0;
+	cdev->private->options.force |= (flags & CCWDEV_ALLOW_FORCE) != 0;
+	return 0;
+}
+
+/**
+ * ccw_device_clear_options() - clear some options
+ * @cdev: device for which the options are to be cleared
+ * @flags: options to be cleared
+ *
+ * All flags specified in @flags are cleared, the remainder is left untouched.
+ */
+void ccw_device_clear_options(struct ccw_device *cdev, unsigned long flags)
+{
+	cdev->private->options.fast &= (flags & CCWDEV_EARLY_NOTIFICATION) == 0;
+	cdev->private->options.repall &= (flags & CCWDEV_REPORT_ALL) == 0;
+	cdev->private->options.pgroup &= (flags & CCWDEV_DO_PATHGROUP) == 0;
+	cdev->private->options.force &= (flags & CCWDEV_ALLOW_FORCE) == 0;
+}
+
+/**
+ * ccw_device_clear() - terminate I/O request processing
+ * @cdev: target ccw device
+ * @intparm: interruption parameter; value is only used if no I/O is
+ *	     outstanding, otherwise the intparm associated with the I/O request
+ *	     is returned
+ *
+ * ccw_device_clear() calls csch on @cdev's subchannel.
+ * Returns:
+ *  %0 on success,
+ *  -%ENODEV on device not operational,
+ *  -%EINVAL on invalid device state.
+ * Context:
+ *  Interrupts disabled, ccw device lock held
+ */
+int ccw_device_clear(struct ccw_device *cdev, unsigned long intparm)
+{
+	struct subchannel *sch;
+	int ret;
+
+	if (!cdev || !cdev->dev.parent)
+		return -ENODEV;
+	sch = to_subchannel(cdev->dev.parent);
+	if (!sch->schib.pmcw.ena)
+		return -EINVAL;
+	if (cdev->private->state == DEV_STATE_NOT_OPER)
+		return -ENODEV;
+	if (cdev->private->state != DEV_STATE_ONLINE &&
+	    cdev->private->state != DEV_STATE_W4SENSE)
+		return -EINVAL;
+
+	ret = cio_clear(sch);
+	if (ret == 0)
+		cdev->private->intparm = intparm;
+	return ret;
+}
+
+/**
+ * ccw_device_start_key() - start a s390 channel program with key
+ * @cdev: target ccw device
+ * @cpa: logical start address of channel program
+ * @intparm: user specific interruption parameter; will be presented back to
+ *	     @cdev's interrupt handler. Allows a device driver to associate
+ *	     the interrupt with a particular I/O request.
+ * @lpm: defines the channel path to be used for a specific I/O request. A
+ *	 value of 0 will make cio use the opm.
+ * @key: storage key to be used for the I/O
+ * @flags: additional flags; defines the action to be performed for I/O
+ *	   processing.
+ *
+ * Start a S/390 channel program. When the interrupt arrives, the
+ * IRQ handler is called, either immediately, delayed (dev-end missing,
+ * or sense required) or never (no IRQ handler registered).
+ * Returns:
+ *  %0, if the operation was successful;
+ *  -%EBUSY, if the device is busy, or status pending;
+ *  -%EACCES, if no path specified in @lpm is operational;
+ *  -%ENODEV, if the device is not operational.
+ * Context:
+ *  Interrupts disabled, ccw device lock held
+ */
+int ccw_device_start_key(struct ccw_device *cdev, struct ccw1 *cpa,
+			 unsigned long intparm, __u8 lpm, __u8 key,
+			 unsigned long flags)
+{
+	struct subchannel *sch;
+	int ret;
+
+	if (!cdev || !cdev->dev.parent)
+		return -ENODEV;
+	sch = to_subchannel(cdev->dev.parent);
+	if (!sch->schib.pmcw.ena)
+		return -EINVAL;
+	if (cdev->private->state == DEV_STATE_NOT_OPER)
+		return -ENODEV;
+	if (cdev->private->state == DEV_STATE_VERIFY ||
+	    cdev->private->state == DEV_STATE_CLEAR_VERIFY) {
+		/* Remember to fake irb when finished. */
+		if (!cdev->private->flags.fake_irb) {
+			cdev->private->flags.fake_irb = 1;
+			cdev->private->intparm = intparm;
+			return 0;
+		} else
+			/* There's already a fake I/O around. */
+			return -EBUSY;
+	}
+	if (cdev->private->state != DEV_STATE_ONLINE ||
+	    ((sch->schib.scsw.cmd.stctl & SCSW_STCTL_PRIM_STATUS) &&
+	     !(sch->schib.scsw.cmd.stctl & SCSW_STCTL_SEC_STATUS)) ||
+	    cdev->private->flags.doverify)
+		return -EBUSY;
+	ret = cio_set_options (sch, flags);
+	if (ret)
+		return ret;
+	/* Adjust requested path mask to excluded varied off paths. */
+	if (lpm) {
+		lpm &= sch->opm;
+		if (lpm == 0)
+			return -EACCES;
+	}
+	ret = cio_start_key (sch, cpa, lpm, key);
+	switch (ret) {
+	case 0:
+		cdev->private->intparm = intparm;
+		break;
+	case -EACCES:
+	case -ENODEV:
+		dev_fsm_event(cdev, DEV_EVENT_VERIFY);
+		break;
+	}
+	return ret;
+}
+
+/**
+ * ccw_device_start_timeout_key() - start a s390 channel program with timeout and key
+ * @cdev: target ccw device
+ * @cpa: logical start address of channel program
+ * @intparm: user specific interruption parameter; will be presented back to
+ *	     @cdev's interrupt handler. Allows a device driver to associate
+ *	     the interrupt with a particular I/O request.
+ * @lpm: defines the channel path to be used for a specific I/O request. A
+ *	 value of 0 will make cio use the opm.
+ * @key: storage key to be used for the I/O
+ * @flags: additional flags; defines the action to be performed for I/O
+ *	   processing.
+ * @expires: timeout value in jiffies
+ *
+ * Start a S/390 channel program. When the interrupt arrives, the
+ * IRQ handler is called, either immediately, delayed (dev-end missing,
+ * or sense required) or never (no IRQ handler registered).
+ * This function notifies the device driver if the channel program has not
+ * completed during the time specified by @expires. If a timeout occurs, the
+ * channel program is terminated via xsch, hsch or csch, and the device's
+ * interrupt handler will be called with an irb containing ERR_PTR(-%ETIMEDOUT).
+ * Returns:
+ *  %0, if the operation was successful;
+ *  -%EBUSY, if the device is busy, or status pending;
+ *  -%EACCES, if no path specified in @lpm is operational;
+ *  -%ENODEV, if the device is not operational.
+ * Context:
+ *  Interrupts disabled, ccw device lock held
+ */
+int ccw_device_start_timeout_key(struct ccw_device *cdev, struct ccw1 *cpa,
+				 unsigned long intparm, __u8 lpm, __u8 key,
+				 unsigned long flags, int expires)
+{
+	int ret;
+
+	if (!cdev)
+		return -ENODEV;
+	ccw_device_set_timeout(cdev, expires);
+	ret = ccw_device_start_key(cdev, cpa, intparm, lpm, key, flags);
+	if (ret != 0)
+		ccw_device_set_timeout(cdev, 0);
+	return ret;
+}
+
+/**
+ * ccw_device_start() - start a s390 channel program
+ * @cdev: target ccw device
+ * @cpa: logical start address of channel program
+ * @intparm: user specific interruption parameter; will be presented back to
+ *	     @cdev's interrupt handler. Allows a device driver to associate
+ *	     the interrupt with a particular I/O request.
+ * @lpm: defines the channel path to be used for a specific I/O request. A
+ *	 value of 0 will make cio use the opm.
+ * @flags: additional flags; defines the action to be performed for I/O
+ *	   processing.
+ *
+ * Start a S/390 channel program. When the interrupt arrives, the
+ * IRQ handler is called, either immediately, delayed (dev-end missing,
+ * or sense required) or never (no IRQ handler registered).
+ * Returns:
+ *  %0, if the operation was successful;
+ *  -%EBUSY, if the device is busy, or status pending;
+ *  -%EACCES, if no path specified in @lpm is operational;
+ *  -%ENODEV, if the device is not operational.
+ * Context:
+ *  Interrupts disabled, ccw device lock held
+ */
+int ccw_device_start(struct ccw_device *cdev, struct ccw1 *cpa,
+		     unsigned long intparm, __u8 lpm, unsigned long flags)
+{
+	return ccw_device_start_key(cdev, cpa, intparm, lpm,
+				    PAGE_DEFAULT_KEY, flags);
+}
+
+/**
+ * ccw_device_start_timeout() - start a s390 channel program with timeout
+ * @cdev: target ccw device
+ * @cpa: logical start address of channel program
+ * @intparm: user specific interruption parameter; will be presented back to
+ *	     @cdev's interrupt handler. Allows a device driver to associate
+ *	     the interrupt with a particular I/O request.
+ * @lpm: defines the channel path to be used for a specific I/O request. A
+ *	 value of 0 will make cio use the opm.
+ * @flags: additional flags; defines the action to be performed for I/O
+ *	   processing.
+ * @expires: timeout value in jiffies
+ *
+ * Start a S/390 channel program. When the interrupt arrives, the
+ * IRQ handler is called, either immediately, delayed (dev-end missing,
+ * or sense required) or never (no IRQ handler registered).
+ * This function notifies the device driver if the channel program has not
+ * completed during the time specified by @expires. If a timeout occurs, the
+ * channel program is terminated via xsch, hsch or csch, and the device's
+ * interrupt handler will be called with an irb containing ERR_PTR(-%ETIMEDOUT).
+ * Returns:
+ *  %0, if the operation was successful;
+ *  -%EBUSY, if the device is busy, or status pending;
+ *  -%EACCES, if no path specified in @lpm is operational;
+ *  -%ENODEV, if the device is not operational.
+ * Context:
+ *  Interrupts disabled, ccw device lock held
+ */
+int ccw_device_start_timeout(struct ccw_device *cdev, struct ccw1 *cpa,
+			     unsigned long intparm, __u8 lpm,
+			     unsigned long flags, int expires)
+{
+	return ccw_device_start_timeout_key(cdev, cpa, intparm, lpm,
+					    PAGE_DEFAULT_KEY, flags,
+					    expires);
+}
+
+
+/**
+ * ccw_device_halt() - halt I/O request processing
+ * @cdev: target ccw device
+ * @intparm: interruption parameter; value is only used if no I/O is
+ *	     outstanding, otherwise the intparm associated with the I/O request
+ *	     is returned
+ *
+ * ccw_device_halt() calls hsch on @cdev's subchannel.
+ * Returns:
+ *  %0 on success,
+ *  -%ENODEV on device not operational,
+ *  -%EINVAL on invalid device state,
+ *  -%EBUSY on device busy or interrupt pending.
+ * Context:
+ *  Interrupts disabled, ccw device lock held
+ */
+int ccw_device_halt(struct ccw_device *cdev, unsigned long intparm)
+{
+	struct subchannel *sch;
+	int ret;
+
+	if (!cdev || !cdev->dev.parent)
+		return -ENODEV;
+	sch = to_subchannel(cdev->dev.parent);
+	if (!sch->schib.pmcw.ena)
+		return -EINVAL;
+	if (cdev->private->state == DEV_STATE_NOT_OPER)
+		return -ENODEV;
+	if (cdev->private->state != DEV_STATE_ONLINE &&
+	    cdev->private->state != DEV_STATE_W4SENSE)
+		return -EINVAL;
+
+	ret = cio_halt(sch);
+	if (ret == 0)
+		cdev->private->intparm = intparm;
+	return ret;
+}
+
+/**
+ * ccw_device_resume() - resume channel program execution
+ * @cdev: target ccw device
+ *
+ * ccw_device_resume() calls rsch on @cdev's subchannel.
+ * Returns:
+ *  %0 on success,
+ *  -%ENODEV on device not operational,
+ *  -%EINVAL on invalid device state,
+ *  -%EBUSY on device busy or interrupt pending.
+ * Context:
+ *  Interrupts disabled, ccw device lock held
+ */
+int ccw_device_resume(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+
+	if (!cdev || !cdev->dev.parent)
+		return -ENODEV;
+	sch = to_subchannel(cdev->dev.parent);
+	if (!sch->schib.pmcw.ena)
+		return -EINVAL;
+	if (cdev->private->state == DEV_STATE_NOT_OPER)
+		return -ENODEV;
+	if (cdev->private->state != DEV_STATE_ONLINE ||
+	    !(sch->schib.scsw.cmd.actl & SCSW_ACTL_SUSPENDED))
+		return -EINVAL;
+	return cio_resume(sch);
+}
+
+/*
+ * Pass interrupt to device driver.
+ */
+int
+ccw_device_call_handler(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	unsigned int stctl;
+	int ending_status;
+
+	sch = to_subchannel(cdev->dev.parent);
+
+	/*
+	 * we allow for the device action handler if .
+	 *  - we received ending status
+	 *  - the action handler requested to see all interrupts
+	 *  - we received an intermediate status
+	 *  - fast notification was requested (primary status)
+	 *  - unsolicited interrupts
+	 */
+	stctl = scsw_stctl(&cdev->private->irb.scsw);
+	ending_status = (stctl & SCSW_STCTL_SEC_STATUS) ||
+		(stctl == (SCSW_STCTL_ALERT_STATUS | SCSW_STCTL_STATUS_PEND)) ||
+		(stctl == SCSW_STCTL_STATUS_PEND);
+	if (!ending_status &&
+	    !cdev->private->options.repall &&
+	    !(stctl & SCSW_STCTL_INTER_STATUS) &&
+	    !(cdev->private->options.fast &&
+	      (stctl & SCSW_STCTL_PRIM_STATUS)))
+		return 0;
+
+	/* Clear pending timers for device driver initiated I/O. */
+	if (ending_status)
+		ccw_device_set_timeout(cdev, 0);
+	/*
+	 * Now we are ready to call the device driver interrupt handler.
+	 */
+	if (cdev->handler)
+		cdev->handler(cdev, cdev->private->intparm,
+			      &cdev->private->irb);
+
+	/*
+	 * Clear the old and now useless interrupt response block.
+	 */
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+
+	return 1;
+}
+
+/**
+ * ccw_device_get_ciw() - Search for CIW command in extended sense data.
+ * @cdev: ccw device to inspect
+ * @ct: command type to look for
+ *
+ * During SenseID, command information words (CIWs) describing special
+ * commands available to the device may have been stored in the extended
+ * sense data. This function searches for CIWs of a specified command
+ * type in the extended sense data.
+ * Returns:
+ *  %NULL if no extended sense data has been stored or if no CIW of the
+ *  specified command type could be found,
+ *  else a pointer to the CIW of the specified command type.
+ */
+struct ciw *ccw_device_get_ciw(struct ccw_device *cdev, __u32 ct)
+{
+	int ciw_cnt;
+
+	if (cdev->private->flags.esid == 0)
+		return NULL;
+	for (ciw_cnt = 0; ciw_cnt < MAX_CIWS; ciw_cnt++)
+		if (cdev->private->senseid.ciw[ciw_cnt].ct == ct)
+			return cdev->private->senseid.ciw + ciw_cnt;
+	return NULL;
+}
+
+/**
+ * ccw_device_get_path_mask() - get currently available paths
+ * @cdev: ccw device to be queried
+ * Returns:
+ *  %0 if no subchannel for the device is available,
+ *  else the mask of currently available paths for the ccw device's subchannel.
+ */
+__u8 ccw_device_get_path_mask(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+
+	if (!cdev->dev.parent)
+		return 0;
+
+	sch = to_subchannel(cdev->dev.parent);
+	return sch->lpm;
+}
+
+/*
+ * Try to break the lock on a boxed device.
+ */
+int
+ccw_device_stlck(struct ccw_device *cdev)
+{
+	void *buf, *buf2;
+	unsigned long flags;
+	struct subchannel *sch;
+	int ret;
+
+	if (!cdev)
+		return -ENODEV;
+
+	if (cdev->drv && !cdev->private->options.force)
+		return -EINVAL;
+
+	sch = to_subchannel(cdev->dev.parent);
+	
+	CIO_TRACE_EVENT(2, "stl lock");
+	CIO_TRACE_EVENT(2, dev_name(&cdev->dev));
+
+	buf = kmalloc(32*sizeof(char), GFP_DMA|GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+	buf2 = kmalloc(32*sizeof(char), GFP_DMA|GFP_KERNEL);
+	if (!buf2) {
+		kfree(buf);
+		return -ENOMEM;
+	}
+	spin_lock_irqsave(sch->lock, flags);
+	ret = cio_enable_subchannel(sch, (u32)(addr_t)sch);
+	if (ret)
+		goto out_unlock;
+	/*
+	 * Setup ccw. We chain an unconditional reserve and a release so we
+	 * only break the lock.
+	 */
+	cdev->private->iccws[0].cmd_code = CCW_CMD_STLCK;
+	cdev->private->iccws[0].cda = (__u32) __pa(buf);
+	cdev->private->iccws[0].count = 32;
+	cdev->private->iccws[0].flags = CCW_FLAG_CC;
+	cdev->private->iccws[1].cmd_code = CCW_CMD_RELEASE;
+	cdev->private->iccws[1].cda = (__u32) __pa(buf2);
+	cdev->private->iccws[1].count = 32;
+	cdev->private->iccws[1].flags = 0;
+	ret = cio_start(sch, cdev->private->iccws, 0);
+	if (ret) {
+		cio_disable_subchannel(sch); //FIXME: return code?
+		goto out_unlock;
+	}
+	cdev->private->irb.scsw.cmd.actl |= SCSW_ACTL_START_PEND;
+	spin_unlock_irqrestore(sch->lock, flags);
+	wait_event(cdev->private->wait_q,
+		   cdev->private->irb.scsw.cmd.actl == 0);
+	spin_lock_irqsave(sch->lock, flags);
+	cio_disable_subchannel(sch); //FIXME: return code?
+	if ((cdev->private->irb.scsw.cmd.dstat !=
+	     (DEV_STAT_CHN_END|DEV_STAT_DEV_END)) ||
+	    (cdev->private->irb.scsw.cmd.cstat != 0))
+		ret = -EIO;
+	/* Clear irb. */
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+out_unlock:
+	kfree(buf);
+	kfree(buf2);
+	spin_unlock_irqrestore(sch->lock, flags);
+	return ret;
+}
+
+void *ccw_device_get_chp_desc(struct ccw_device *cdev, int chp_no)
+{
+	struct subchannel *sch;
+	struct chp_id chpid;
+
+	sch = to_subchannel(cdev->dev.parent);
+	chp_id_init(&chpid);
+	chpid.id = sch->schib.pmcw.chpid[chp_no];
+	return chp_get_chp_desc(chpid);
+}
+
+/**
+ * ccw_device_get_id - obtain a ccw device id
+ * @cdev: device to obtain the id for
+ * @dev_id: where to fill in the values
+ */
+void ccw_device_get_id(struct ccw_device *cdev, struct ccw_dev_id *dev_id)
+{
+	*dev_id = cdev->private->dev_id;
+}
+EXPORT_SYMBOL(ccw_device_get_id);
+
+/**
+ * ccw_device_tm_start_key - perform start function
+ * @cdev: ccw device on which to perform the start function
+ * @tcw: transport-command word to be started
+ * @intparm: user defined parameter to be passed to the interrupt handler
+ * @lpm: mask of paths to use
+ * @key: storage key to use for storage access
+ *
+ * Start the tcw on the given ccw device. Return zero on success, non-zero
+ * otherwise.
+ */
+int ccw_device_tm_start_key(struct ccw_device *cdev, struct tcw *tcw,
+			    unsigned long intparm, u8 lpm, u8 key)
+{
+	struct subchannel *sch;
+	int rc;
+
+	sch = to_subchannel(cdev->dev.parent);
+	if (!sch->schib.pmcw.ena)
+		return -EINVAL;
+	if (cdev->private->state != DEV_STATE_ONLINE)
+		return -EIO;
+	/* Adjust requested path mask to excluded varied off paths. */
+	if (lpm) {
+		lpm &= sch->opm;
+		if (lpm == 0)
+			return -EACCES;
+	}
+	rc = cio_tm_start_key(sch, tcw, lpm, key);
+	if (rc == 0)
+		cdev->private->intparm = intparm;
+	return rc;
+}
+EXPORT_SYMBOL(ccw_device_tm_start_key);
+
+/**
+ * ccw_device_tm_start_timeout_key - perform start function
+ * @cdev: ccw device on which to perform the start function
+ * @tcw: transport-command word to be started
+ * @intparm: user defined parameter to be passed to the interrupt handler
+ * @lpm: mask of paths to use
+ * @key: storage key to use for storage access
+ * @expires: time span in jiffies after which to abort request
+ *
+ * Start the tcw on the given ccw device. Return zero on success, non-zero
+ * otherwise.
+ */
+int ccw_device_tm_start_timeout_key(struct ccw_device *cdev, struct tcw *tcw,
+				    unsigned long intparm, u8 lpm, u8 key,
+				    int expires)
+{
+	int ret;
+
+	ccw_device_set_timeout(cdev, expires);
+	ret = ccw_device_tm_start_key(cdev, tcw, intparm, lpm, key);
+	if (ret != 0)
+		ccw_device_set_timeout(cdev, 0);
+	return ret;
+}
+EXPORT_SYMBOL(ccw_device_tm_start_timeout_key);
+
+/**
+ * ccw_device_tm_start - perform start function
+ * @cdev: ccw device on which to perform the start function
+ * @tcw: transport-command word to be started
+ * @intparm: user defined parameter to be passed to the interrupt handler
+ * @lpm: mask of paths to use
+ *
+ * Start the tcw on the given ccw device. Return zero on success, non-zero
+ * otherwise.
+ */
+int ccw_device_tm_start(struct ccw_device *cdev, struct tcw *tcw,
+			unsigned long intparm, u8 lpm)
+{
+	return ccw_device_tm_start_key(cdev, tcw, intparm, lpm,
+				       PAGE_DEFAULT_KEY);
+}
+EXPORT_SYMBOL(ccw_device_tm_start);
+
+/**
+ * ccw_device_tm_start_timeout - perform start function
+ * @cdev: ccw device on which to perform the start function
+ * @tcw: transport-command word to be started
+ * @intparm: user defined parameter to be passed to the interrupt handler
+ * @lpm: mask of paths to use
+ * @expires: time span in jiffies after which to abort request
+ *
+ * Start the tcw on the given ccw device. Return zero on success, non-zero
+ * otherwise.
+ */
+int ccw_device_tm_start_timeout(struct ccw_device *cdev, struct tcw *tcw,
+			       unsigned long intparm, u8 lpm, int expires)
+{
+	return ccw_device_tm_start_timeout_key(cdev, tcw, intparm, lpm,
+					       PAGE_DEFAULT_KEY, expires);
+}
+EXPORT_SYMBOL(ccw_device_tm_start_timeout);
+
+/**
+ * ccw_device_tm_intrg - perform interrogate function
+ * @cdev: ccw device on which to perform the interrogate function
+ *
+ * Perform an interrogate function on the given ccw device. Return zero on
+ * success, non-zero otherwise.
+ */
+int ccw_device_tm_intrg(struct ccw_device *cdev)
+{
+	struct subchannel *sch = to_subchannel(cdev->dev.parent);
+
+	if (!sch->schib.pmcw.ena)
+		return -EINVAL;
+	if (cdev->private->state != DEV_STATE_ONLINE)
+		return -EIO;
+	if (!scsw_is_tm(&sch->schib.scsw) ||
+	    !(scsw_actl(&sch->schib.scsw) & SCSW_ACTL_START_PEND))
+		return -EINVAL;
+	return cio_tm_intrg(sch);
+}
+EXPORT_SYMBOL(ccw_device_tm_intrg);
+
+// FIXME: these have to go:
+
+int
+_ccw_device_get_subchannel_number(struct ccw_device *cdev)
+{
+	return cdev->private->schid.sch_no;
+}
+
+
+MODULE_LICENSE("GPL");
+EXPORT_SYMBOL(ccw_device_set_options_mask);
+EXPORT_SYMBOL(ccw_device_set_options);
+EXPORT_SYMBOL(ccw_device_clear_options);
+EXPORT_SYMBOL(ccw_device_clear);
+EXPORT_SYMBOL(ccw_device_halt);
+EXPORT_SYMBOL(ccw_device_resume);
+EXPORT_SYMBOL(ccw_device_start_timeout);
+EXPORT_SYMBOL(ccw_device_start);
+EXPORT_SYMBOL(ccw_device_start_timeout_key);
+EXPORT_SYMBOL(ccw_device_start_key);
+EXPORT_SYMBOL(ccw_device_get_ciw);
+EXPORT_SYMBOL(ccw_device_get_path_mask);
+EXPORT_SYMBOL(_ccw_device_get_subchannel_number);
+EXPORT_SYMBOL_GPL(ccw_device_get_chp_desc);

kernel/drivers/s390/cio/device_pgid.c

@@ -0,0 +1,594 @@
+/*
+ * drivers/s390/cio/device_pgid.c
+ *
+ *    Copyright (C) 2002 IBM Deutschland Entwicklung GmbH,
+ *			 IBM Corporation
+ *    Author(s): Cornelia Huck (cornelia.huck@de.ibm.com)
+ *		 Martin Schwidefsky (schwidefsky@de.ibm.com)
+ *
+ * Path Group ID functions.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+
+#include <asm/ccwdev.h>
+#include <asm/cio.h>
+#include <asm/delay.h>
+#include <asm/lowcore.h>
+
+#include "cio.h"
+#include "cio_debug.h"
+#include "css.h"
+#include "device.h"
+#include "ioasm.h"
+#include "io_sch.h"
+
+/*
+ * Helper function called from interrupt context to decide whether an
+ * operation should be tried again.
+ */
+static int __ccw_device_should_retry(union scsw *scsw)
+{
+	/* CC is only valid if start function bit is set. */
+	if ((scsw->cmd.fctl & SCSW_FCTL_START_FUNC) && scsw->cmd.cc == 1)
+		return 1;
+	/* No more activity. For sense and set PGID we stubbornly try again. */
+	if (!scsw->cmd.actl)
+		return 1;
+	return 0;
+}
+
+/*
+ * Start Sense Path Group ID helper function. Used in ccw_device_recog
+ * and ccw_device_sense_pgid.
+ */
+static int
+__ccw_device_sense_pgid_start(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	struct ccw1 *ccw;
+	int ret;
+	int i;
+
+	sch = to_subchannel(cdev->dev.parent);
+	/* Return if we already checked on all paths. */
+	if (cdev->private->imask == 0)
+		return (sch->lpm == 0) ? -ENODEV : -EACCES;
+	i = 8 - ffs(cdev->private->imask);
+
+	/* Setup sense path group id channel program. */
+	ccw = cdev->private->iccws;
+	ccw->cmd_code = CCW_CMD_SENSE_PGID;
+	ccw->count = sizeof (struct pgid);
+	ccw->flags = CCW_FLAG_SLI;
+
+	/* Reset device status. */
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+	/* Try on every path. */
+	ret = -ENODEV;
+	while (cdev->private->imask != 0) {
+		/* Try every path multiple times. */
+		ccw->cda = (__u32) __pa (&cdev->private->pgid[i]);
+		if (cdev->private->iretry > 0) {
+			cdev->private->iretry--;
+			/* Reset internal retry indication. */
+			cdev->private->flags.intretry = 0;
+			ret = cio_start (sch, cdev->private->iccws, 
+					 cdev->private->imask);
+			/* ret is 0, -EBUSY, -EACCES or -ENODEV */
+			if (ret != -EACCES)
+				return ret;
+			CIO_MSG_EVENT(3, "SNID - Device %04x on Subchannel "
+				      "0.%x.%04x, lpm %02X, became 'not "
+				      "operational'\n",
+				      cdev->private->dev_id.devno,
+				      sch->schid.ssid,
+				      sch->schid.sch_no, cdev->private->imask);
+
+		}
+		cdev->private->imask >>= 1;
+		cdev->private->iretry = 5;
+		i++;
+	}
+
+	return ret;
+}
+
+void
+ccw_device_sense_pgid_start(struct ccw_device *cdev)
+{
+	int ret;
+
+	/* Set a timeout of 60s */
+	ccw_device_set_timeout(cdev, 60*HZ);
+
+	cdev->private->state = DEV_STATE_SENSE_PGID;
+	cdev->private->imask = 0x80;
+	cdev->private->iretry = 5;
+	memset (&cdev->private->pgid, 0, sizeof (cdev->private->pgid));
+	ret = __ccw_device_sense_pgid_start(cdev);
+	if (ret && ret != -EBUSY)
+		ccw_device_sense_pgid_done(cdev, ret);
+}
+
+/*
+ * Called from interrupt context to check if a valid answer
+ * to Sense Path Group ID was received.
+ */
+static int
+__ccw_device_check_sense_pgid(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	struct irb *irb;
+	int i;
+
+	sch = to_subchannel(cdev->dev.parent);
+	irb = &cdev->private->irb;
+	if (irb->scsw.cmd.fctl & (SCSW_FCTL_HALT_FUNC | SCSW_FCTL_CLEAR_FUNC)) {
+		/* Retry Sense PGID if requested. */
+		if (cdev->private->flags.intretry) {
+			cdev->private->flags.intretry = 0;
+			return -EAGAIN;
+		}
+		return -ETIME;
+	}
+	if (irb->esw.esw0.erw.cons &&
+	    (irb->ecw[0]&(SNS0_CMD_REJECT|SNS0_INTERVENTION_REQ))) {
+		/*
+		 * If the device doesn't support the Sense Path Group ID
+		 *  command further retries wouldn't help ...
+		 */
+		return -EOPNOTSUPP;
+	}
+	if (irb->esw.esw0.erw.cons) {
+		CIO_MSG_EVENT(2, "SNID - device 0.%x.%04x, unit check, "
+			      "lpum %02X, cnt %02d, sns : "
+			      "%02X%02X%02X%02X %02X%02X%02X%02X ...\n",
+			      cdev->private->dev_id.ssid,
+			      cdev->private->dev_id.devno,
+			      irb->esw.esw0.sublog.lpum,
+			      irb->esw.esw0.erw.scnt,
+			      irb->ecw[0], irb->ecw[1],
+			      irb->ecw[2], irb->ecw[3],
+			      irb->ecw[4], irb->ecw[5],
+			      irb->ecw[6], irb->ecw[7]);
+		return -EAGAIN;
+	}
+	if (irb->scsw.cmd.cc == 3) {
+		u8 lpm;
+
+		lpm = to_io_private(sch)->orb.cmd.lpm;
+		CIO_MSG_EVENT(3, "SNID - Device %04x on Subchannel 0.%x.%04x,"
+			      " lpm %02X, became 'not operational'\n",
+			      cdev->private->dev_id.devno, sch->schid.ssid,
+			      sch->schid.sch_no, lpm);
+		return -EACCES;
+	}
+	i = 8 - ffs(cdev->private->imask);
+	if (cdev->private->pgid[i].inf.ps.state2 == SNID_STATE2_RESVD_ELSE) {
+		CIO_MSG_EVENT(2, "SNID - Device %04x on Subchannel 0.%x.%04x "
+			      "is reserved by someone else\n",
+			      cdev->private->dev_id.devno, sch->schid.ssid,
+			      sch->schid.sch_no);
+		return -EUSERS;
+	}
+	return 0;
+}
+
+/*
+ * Got interrupt for Sense Path Group ID.
+ */
+void
+ccw_device_sense_pgid_irq(struct ccw_device *cdev, enum dev_event dev_event)
+{
+	struct subchannel *sch;
+	struct irb *irb;
+	int ret;
+
+	irb = (struct irb *) __LC_IRB;
+
+	if (irb->scsw.cmd.stctl ==
+	    (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)) {
+		if (__ccw_device_should_retry(&irb->scsw)) {
+			ret = __ccw_device_sense_pgid_start(cdev);
+			if (ret && ret != -EBUSY)
+				ccw_device_sense_pgid_done(cdev, ret);
+		}
+		return;
+	}
+	if (ccw_device_accumulate_and_sense(cdev, irb) != 0)
+		return;
+	sch = to_subchannel(cdev->dev.parent);
+	ret = __ccw_device_check_sense_pgid(cdev);
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+	switch (ret) {
+	/* 0, -ETIME, -EOPNOTSUPP, -EAGAIN, -EACCES or -EUSERS */
+	case -EOPNOTSUPP:	/* Sense Path Group ID not supported */
+		ccw_device_sense_pgid_done(cdev, -EOPNOTSUPP);
+		break;
+	case -ETIME:		/* Sense path group id stopped by timeout. */
+		ccw_device_sense_pgid_done(cdev, -ETIME);
+		break;
+	case -EACCES:		/* channel is not operational. */
+		sch->lpm &= ~cdev->private->imask;
+		/* Fall through. */
+	case 0:			/* Sense Path Group ID successful. */
+		cdev->private->imask >>= 1;
+		cdev->private->iretry = 5;
+		/* Fall through. */
+	case -EAGAIN:		/* Try again. */
+		ret = __ccw_device_sense_pgid_start(cdev);
+		if (ret != 0 && ret != -EBUSY)
+			ccw_device_sense_pgid_done(cdev, ret);
+		break;
+	case -EUSERS:		/* device is reserved for someone else. */
+		ccw_device_sense_pgid_done(cdev, -EUSERS);
+		break;
+	}
+}
+
+/*
+ * Path Group ID helper function.
+ */
+static int
+__ccw_device_do_pgid(struct ccw_device *cdev, __u8 func)
+{
+	struct subchannel *sch;
+	struct ccw1 *ccw;
+	int ret;
+
+	sch = to_subchannel(cdev->dev.parent);
+
+	/* Setup sense path group id channel program. */
+	cdev->private->pgid[0].inf.fc = func;
+	ccw = cdev->private->iccws;
+	if (cdev->private->flags.pgid_single)
+		cdev->private->pgid[0].inf.fc |= SPID_FUNC_SINGLE_PATH;
+	else
+		cdev->private->pgid[0].inf.fc |= SPID_FUNC_MULTI_PATH;
+	ccw->cmd_code = CCW_CMD_SET_PGID;
+	ccw->cda = (__u32) __pa (&cdev->private->pgid[0]);
+	ccw->count = sizeof (struct pgid);
+	ccw->flags = CCW_FLAG_SLI;
+
+	/* Reset device status. */
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+
+	/* Try multiple times. */
+	ret = -EACCES;
+	if (cdev->private->iretry > 0) {
+		cdev->private->iretry--;
+		/* Reset internal retry indication. */
+		cdev->private->flags.intretry = 0;
+		ret = cio_start (sch, cdev->private->iccws,
+				 cdev->private->imask);
+		/* We expect an interrupt in case of success or busy
+		 * indication. */
+		if ((ret == 0) || (ret == -EBUSY))
+			return ret;
+	}
+	/* PGID command failed on this path. */
+	CIO_MSG_EVENT(3, "SPID - Device %04x on Subchannel "
+		      "0.%x.%04x, lpm %02X, became 'not operational'\n",
+		      cdev->private->dev_id.devno, sch->schid.ssid,
+		      sch->schid.sch_no, cdev->private->imask);
+	return ret;
+}
+
+/*
+ * Helper function to send a nop ccw down a path.
+ */
+static int __ccw_device_do_nop(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	struct ccw1 *ccw;
+	int ret;
+
+	sch = to_subchannel(cdev->dev.parent);
+
+	/* Setup nop channel program. */
+	ccw = cdev->private->iccws;
+	ccw->cmd_code = CCW_CMD_NOOP;
+	ccw->cda = 0;
+	ccw->count = 0;
+	ccw->flags = CCW_FLAG_SLI;
+
+	/* Reset device status. */
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+
+	/* Try multiple times. */
+	ret = -EACCES;
+	if (cdev->private->iretry > 0) {
+		cdev->private->iretry--;
+		/* Reset internal retry indication. */
+		cdev->private->flags.intretry = 0;
+		ret = cio_start (sch, cdev->private->iccws,
+				 cdev->private->imask);
+		/* We expect an interrupt in case of success or busy
+		 * indication. */
+		if ((ret == 0) || (ret == -EBUSY))
+			return ret;
+	}
+	/* nop command failed on this path. */
+	CIO_MSG_EVENT(3, "NOP - Device %04x on Subchannel "
+		      "0.%x.%04x, lpm %02X, became 'not operational'\n",
+		      cdev->private->dev_id.devno, sch->schid.ssid,
+		      sch->schid.sch_no, cdev->private->imask);
+	return ret;
+}
+
+
+/*
+ * Called from interrupt context to check if a valid answer
+ * to Set Path Group ID was received.
+ */
+static int
+__ccw_device_check_pgid(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	struct irb *irb;
+
+	sch = to_subchannel(cdev->dev.parent);
+	irb = &cdev->private->irb;
+	if (irb->scsw.cmd.fctl & (SCSW_FCTL_HALT_FUNC | SCSW_FCTL_CLEAR_FUNC)) {
+		/* Retry Set PGID if requested. */
+		if (cdev->private->flags.intretry) {
+			cdev->private->flags.intretry = 0;
+			return -EAGAIN;
+		}
+		return -ETIME;
+	}
+	if (irb->esw.esw0.erw.cons) {
+		if (irb->ecw[0] & SNS0_CMD_REJECT)
+			return -EOPNOTSUPP;
+		/* Hmm, whatever happened, try again. */
+		CIO_MSG_EVENT(2, "SPID - device 0.%x.%04x, unit check, "
+			      "cnt %02d, "
+			      "sns : %02X%02X%02X%02X %02X%02X%02X%02X ...\n",
+			      cdev->private->dev_id.ssid,
+			      cdev->private->dev_id.devno,
+			      irb->esw.esw0.erw.scnt,
+			      irb->ecw[0], irb->ecw[1],
+			      irb->ecw[2], irb->ecw[3],
+			      irb->ecw[4], irb->ecw[5],
+			      irb->ecw[6], irb->ecw[7]);
+		return -EAGAIN;
+	}
+	if (irb->scsw.cmd.cc == 3) {
+		CIO_MSG_EVENT(3, "SPID - Device %04x on Subchannel 0.%x.%04x,"
+			      " lpm %02X, became 'not operational'\n",
+			      cdev->private->dev_id.devno, sch->schid.ssid,
+			      sch->schid.sch_no, cdev->private->imask);
+		return -EACCES;
+	}
+	return 0;
+}
+
+/*
+ * Called from interrupt context to check the path status after a nop has
+ * been send.
+ */
+static int __ccw_device_check_nop(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	struct irb *irb;
+
+	sch = to_subchannel(cdev->dev.parent);
+	irb = &cdev->private->irb;
+	if (irb->scsw.cmd.fctl & (SCSW_FCTL_HALT_FUNC | SCSW_FCTL_CLEAR_FUNC)) {
+		/* Retry NOP if requested. */
+		if (cdev->private->flags.intretry) {
+			cdev->private->flags.intretry = 0;
+			return -EAGAIN;
+		}
+		return -ETIME;
+	}
+	if (irb->scsw.cmd.cc == 3) {
+		CIO_MSG_EVENT(3, "NOP - Device %04x on Subchannel 0.%x.%04x,"
+			      " lpm %02X, became 'not operational'\n",
+			      cdev->private->dev_id.devno, sch->schid.ssid,
+			      sch->schid.sch_no, cdev->private->imask);
+		return -EACCES;
+	}
+	return 0;
+}
+
+static void
+__ccw_device_verify_start(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	__u8 func;
+	int ret;
+
+	sch = to_subchannel(cdev->dev.parent);
+	/* Repeat for all paths. */
+	for (; cdev->private->imask; cdev->private->imask >>= 1,
+				     cdev->private->iretry = 5) {
+		if ((cdev->private->imask & sch->schib.pmcw.pam) == 0)
+			/* Path not available, try next. */
+			continue;
+		if (cdev->private->options.pgroup) {
+			if (sch->opm & cdev->private->imask)
+				func = SPID_FUNC_ESTABLISH;
+			else
+				func = SPID_FUNC_RESIGN;
+			ret = __ccw_device_do_pgid(cdev, func);
+		} else
+			ret = __ccw_device_do_nop(cdev);
+		/* We expect an interrupt in case of success or busy
+		 * indication. */
+		if (ret == 0 || ret == -EBUSY)
+			return;
+		/* Permanent path failure, try next. */
+	}
+	/* Done with all paths. */
+	ccw_device_verify_done(cdev, (sch->vpm != 0) ? 0 : -ENODEV);
+}
+		
+/*
+ * Got interrupt for Set Path Group ID.
+ */
+void
+ccw_device_verify_irq(struct ccw_device *cdev, enum dev_event dev_event)
+{
+	struct subchannel *sch;
+	struct irb *irb;
+	int ret;
+
+	irb = (struct irb *) __LC_IRB;
+
+	if (irb->scsw.cmd.stctl ==
+	    (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)) {
+		if (__ccw_device_should_retry(&irb->scsw))
+			__ccw_device_verify_start(cdev);
+		return;
+	}
+	if (ccw_device_accumulate_and_sense(cdev, irb) != 0)
+		return;
+	sch = to_subchannel(cdev->dev.parent);
+	if (cdev->private->options.pgroup)
+		ret = __ccw_device_check_pgid(cdev);
+	else
+		ret = __ccw_device_check_nop(cdev);
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+
+	switch (ret) {
+	/* 0, -ETIME, -EAGAIN, -EOPNOTSUPP or -EACCES */
+	case 0:
+		/* Path verification ccw finished successfully, update lpm. */
+		sch->vpm |= sch->opm & cdev->private->imask;
+		/* Go on with next path. */
+		cdev->private->imask >>= 1;
+		cdev->private->iretry = 5;
+		__ccw_device_verify_start(cdev);
+		break;
+	case -EOPNOTSUPP:
+		/*
+		 * One of those strange devices which claim to be able
+		 * to do multipathing but not for Set Path Group ID.
+		 */
+		if (cdev->private->flags.pgid_single)
+			cdev->private->options.pgroup = 0;
+		else
+			cdev->private->flags.pgid_single = 1;
+		/* Retry */
+		sch->vpm = 0;
+		cdev->private->imask = 0x80;
+		cdev->private->iretry = 5;
+		/* fall through. */
+	case -EAGAIN:		/* Try again. */
+		__ccw_device_verify_start(cdev);
+		break;
+	case -ETIME:		/* Set path group id stopped by timeout. */
+		ccw_device_verify_done(cdev, -ETIME);
+		break;
+	case -EACCES:		/* channel is not operational. */
+		cdev->private->imask >>= 1;
+		cdev->private->iretry = 5;
+		__ccw_device_verify_start(cdev);
+		break;
+	}
+}
+
+void
+ccw_device_verify_start(struct ccw_device *cdev)
+{
+	struct subchannel *sch = to_subchannel(cdev->dev.parent);
+
+	cdev->private->flags.pgid_single = 0;
+	cdev->private->imask = 0x80;
+	cdev->private->iretry = 5;
+
+	/* Start with empty vpm. */
+	sch->vpm = 0;
+
+	/* Get current pam. */
+	if (cio_update_schib(sch)) {
+		ccw_device_verify_done(cdev, -ENODEV);
+		return;
+	}
+	/* After 60s path verification is considered to have failed. */
+	ccw_device_set_timeout(cdev, 60*HZ);
+	__ccw_device_verify_start(cdev);
+}
+
+static void
+__ccw_device_disband_start(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+	int ret;
+
+	sch = to_subchannel(cdev->dev.parent);
+	while (cdev->private->imask != 0) {
+		if (sch->lpm & cdev->private->imask) {
+			ret = __ccw_device_do_pgid(cdev, SPID_FUNC_DISBAND);
+			if (ret == 0)
+				return;
+		}
+		cdev->private->iretry = 5;
+		cdev->private->imask >>= 1;
+	}
+	ccw_device_disband_done(cdev, (sch->lpm != 0) ? 0 : -ENODEV);
+}
+
+/*
+ * Got interrupt for Unset Path Group ID.
+ */
+void
+ccw_device_disband_irq(struct ccw_device *cdev, enum dev_event dev_event)
+{
+	struct subchannel *sch;
+	struct irb *irb;
+	int ret;
+
+	irb = (struct irb *) __LC_IRB;
+
+	if (irb->scsw.cmd.stctl ==
+	    (SCSW_STCTL_STATUS_PEND | SCSW_STCTL_ALERT_STATUS)) {
+		if (__ccw_device_should_retry(&irb->scsw))
+			__ccw_device_disband_start(cdev);
+		return;
+	}
+	if (ccw_device_accumulate_and_sense(cdev, irb) != 0)
+		return;
+	sch = to_subchannel(cdev->dev.parent);
+	ret = __ccw_device_check_pgid(cdev);
+	memset(&cdev->private->irb, 0, sizeof(struct irb));
+	switch (ret) {
+	/* 0, -ETIME, -EAGAIN, -EOPNOTSUPP or -EACCES */
+	case 0:			/* disband successful. */
+		ccw_device_disband_done(cdev, ret);
+		break;
+	case -EOPNOTSUPP:
+		/*
+		 * One of those strange devices which claim to be able
+		 * to do multipathing but not for Unset Path Group ID.
+		 */
+		cdev->private->flags.pgid_single = 1;
+		/* fall through. */
+	case -EAGAIN:		/* Try again. */
+		__ccw_device_disband_start(cdev);
+		break;
+	case -ETIME:		/* Set path group id stopped by timeout. */
+		ccw_device_disband_done(cdev, -ETIME);
+		break;
+	case -EACCES:		/* channel is not operational. */
+		cdev->private->imask >>= 1;
+		cdev->private->iretry = 5;
+		__ccw_device_disband_start(cdev);
+		break;
+	}
+}
+
+void
+ccw_device_disband_start(struct ccw_device *cdev)
+{
+	/* After 60s disbanding is considered to have failed. */
+	ccw_device_set_timeout(cdev, 60*HZ);
+
+	cdev->private->flags.pgid_single = 0;
+	cdev->private->iretry = 5;
+	cdev->private->imask = 0x80;
+	__ccw_device_disband_start(cdev);
+}

kernel/drivers/s390/cio/device_status.c

@@ -0,0 +1,403 @@
+/*
+ * drivers/s390/cio/device_status.c
+ *
+ *    Copyright (C) 2002 IBM Deutschland Entwicklung GmbH,
+ *			 IBM Corporation
+ *    Author(s): Cornelia Huck (cornelia.huck@de.ibm.com)
+ *		 Martin Schwidefsky (schwidefsky@de.ibm.com)
+ *
+ * Status accumulation and basic sense functions.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+
+#include <asm/ccwdev.h>
+#include <asm/cio.h>
+
+#include "cio.h"
+#include "cio_debug.h"
+#include "css.h"
+#include "device.h"
+#include "ioasm.h"
+#include "io_sch.h"
+
+/*
+ * Check for any kind of channel or interface control check but don't
+ * issue the message for the console device
+ */
+static void
+ccw_device_msg_control_check(struct ccw_device *cdev, struct irb *irb)
+{
+	char dbf_text[15];
+
+	if (!scsw_is_valid_cstat(&irb->scsw) ||
+	    !(scsw_cstat(&irb->scsw) & (SCHN_STAT_CHN_DATA_CHK |
+	      SCHN_STAT_CHN_CTRL_CHK | SCHN_STAT_INTF_CTRL_CHK)))
+		return;
+	CIO_MSG_EVENT(0, "Channel-Check or Interface-Control-Check "
+		      "received"
+		      " ... device %04x on subchannel 0.%x.%04x, dev_stat "
+		      ": %02X sch_stat : %02X\n",
+		      cdev->private->dev_id.devno, cdev->private->schid.ssid,
+		      cdev->private->schid.sch_no,
+		      scsw_dstat(&irb->scsw), scsw_cstat(&irb->scsw));
+	sprintf(dbf_text, "chk%x", cdev->private->schid.sch_no);
+	CIO_TRACE_EVENT(0, dbf_text);
+	CIO_HEX_EVENT(0, irb, sizeof(struct irb));
+}
+
+/*
+ * Some paths became not operational (pno bit in scsw is set).
+ */
+static void
+ccw_device_path_notoper(struct ccw_device *cdev)
+{
+	struct subchannel *sch;
+
+	sch = to_subchannel(cdev->dev.parent);
+	if (cio_update_schib(sch))
+		goto doverify;
+
+	CIO_MSG_EVENT(0, "%s(0.%x.%04x) - path(s) %02x are "
+		      "not operational \n", __func__,
+		      sch->schid.ssid, sch->schid.sch_no,
+		      sch->schib.pmcw.pnom);
+
+	sch->lpm &= ~sch->schib.pmcw.pnom;
+doverify:
+	cdev->private->flags.doverify = 1;
+}
+
+/*
+ * Copy valid bits from the extended control word to device irb.
+ */
+static void
+ccw_device_accumulate_ecw(struct ccw_device *cdev, struct irb *irb)
+{
+	/*
+	 * Copy extended control bit if it is valid... yes there
+	 * are condition that have to be met for the extended control
+	 * bit to have meaning. Sick.
+	 */
+	cdev->private->irb.scsw.cmd.ectl = 0;
+	if ((irb->scsw.cmd.stctl & SCSW_STCTL_ALERT_STATUS) &&
+	    !(irb->scsw.cmd.stctl & SCSW_STCTL_INTER_STATUS))
+		cdev->private->irb.scsw.cmd.ectl = irb->scsw.cmd.ectl;
+	/* Check if extended control word is valid. */
+	if (!cdev->private->irb.scsw.cmd.ectl)
+		return;
+	/* Copy concurrent sense / model dependent information. */
+	memcpy (&cdev->private->irb.ecw, irb->ecw, sizeof (irb->ecw));
+}
+
+/*
+ * Check if extended status word is valid.
+ */
+static int
+ccw_device_accumulate_esw_valid(struct irb *irb)
+{
+	if (!irb->scsw.cmd.eswf &&
+	    (irb->scsw.cmd.stctl == SCSW_STCTL_STATUS_PEND))
+		return 0;
+	if (irb->scsw.cmd.stctl ==
+			(SCSW_STCTL_INTER_STATUS|SCSW_STCTL_STATUS_PEND) &&
+	    !(irb->scsw.cmd.actl & SCSW_ACTL_SUSPENDED))
+		return 0;
+	return 1;
+}
+
+/*
+ * Copy valid bits from the extended status word to device irb.
+ */
+static void
+ccw_device_accumulate_esw(struct ccw_device *cdev, struct irb *irb)
+{
+	struct irb *cdev_irb;
+	struct sublog *cdev_sublog, *sublog;
+
+	if (!ccw_device_accumulate_esw_valid(irb))
+		return;
+
+	cdev_irb = &cdev->private->irb;
+
+	/* Copy last path used mask. */
+	cdev_irb->esw.esw1.lpum = irb->esw.esw1.lpum;
+
+	/* Copy subchannel logout information if esw is of format 0. */
+	if (irb->scsw.cmd.eswf) {
+		cdev_sublog = &cdev_irb->esw.esw0.sublog;
+		sublog = &irb->esw.esw0.sublog;
+		/* Copy extended status flags. */
+		cdev_sublog->esf = sublog->esf;
+		/*
+		 * Copy fields that have a meaning for channel data check
+		 * channel control check and interface control check.
+		 */
+		if (irb->scsw.cmd.cstat & (SCHN_STAT_CHN_DATA_CHK |
+				       SCHN_STAT_CHN_CTRL_CHK |
+				       SCHN_STAT_INTF_CTRL_CHK)) {
+			/* Copy ancillary report bit. */
+			cdev_sublog->arep = sublog->arep;
+			/* Copy field-validity-flags. */
+			cdev_sublog->fvf = sublog->fvf;
+			/* Copy storage access code. */
+			cdev_sublog->sacc = sublog->sacc;
+			/* Copy termination code. */
+			cdev_sublog->termc = sublog->termc;
+			/* Copy sequence code. */
+			cdev_sublog->seqc = sublog->seqc;
+		}
+		/* Copy device status check. */
+		cdev_sublog->devsc = sublog->devsc;
+		/* Copy secondary error. */
+		cdev_sublog->serr = sublog->serr;
+		/* Copy i/o-error alert. */
+		cdev_sublog->ioerr = sublog->ioerr;
+		/* Copy channel path timeout bit. */
+		if (irb->scsw.cmd.cstat & SCHN_STAT_INTF_CTRL_CHK)
+			cdev_irb->esw.esw0.erw.cpt = irb->esw.esw0.erw.cpt;
+		/* Copy failing storage address validity flag. */
+		cdev_irb->esw.esw0.erw.fsavf = irb->esw.esw0.erw.fsavf;
+		if (cdev_irb->esw.esw0.erw.fsavf) {
+			/* ... and copy the failing storage address. */
+			memcpy(cdev_irb->esw.esw0.faddr, irb->esw.esw0.faddr,
+			       sizeof (irb->esw.esw0.faddr));
+			/* ... and copy the failing storage address format. */
+			cdev_irb->esw.esw0.erw.fsaf = irb->esw.esw0.erw.fsaf;
+		}
+		/* Copy secondary ccw address validity bit. */
+		cdev_irb->esw.esw0.erw.scavf = irb->esw.esw0.erw.scavf;
+		if (irb->esw.esw0.erw.scavf)
+			/* ... and copy the secondary ccw address. */
+			cdev_irb->esw.esw0.saddr = irb->esw.esw0.saddr;
+		
+	}
+	/* FIXME: DCTI for format 2? */
+
+	/* Copy authorization bit. */
+	cdev_irb->esw.esw0.erw.auth = irb->esw.esw0.erw.auth;
+	/* Copy path verification required flag. */
+	cdev_irb->esw.esw0.erw.pvrf = irb->esw.esw0.erw.pvrf;
+	if (irb->esw.esw0.erw.pvrf)
+		cdev->private->flags.doverify = 1;
+	/* Copy concurrent sense bit. */
+	cdev_irb->esw.esw0.erw.cons = irb->esw.esw0.erw.cons;
+	if (irb->esw.esw0.erw.cons)
+		cdev_irb->esw.esw0.erw.scnt = irb->esw.esw0.erw.scnt;
+}
+
+/*
+ * Accumulate status from irb to devstat.
+ */
+void
+ccw_device_accumulate_irb(struct ccw_device *cdev, struct irb *irb)
+{
+	struct irb *cdev_irb;
+
+	/*
+	 * Check if the status pending bit is set in stctl.
+	 * If not, the remaining bit have no meaning and we must ignore them.
+	 * The esw is not meaningful as well...
+	 */
+	if (!(scsw_stctl(&irb->scsw) & SCSW_STCTL_STATUS_PEND))
+		return;
+
+	/* Check for channel checks and interface control checks. */
+	ccw_device_msg_control_check(cdev, irb);
+
+	/* Check for path not operational. */
+	if (scsw_is_valid_pno(&irb->scsw) && scsw_pno(&irb->scsw))
+		ccw_device_path_notoper(cdev);
+	/* No irb accumulation for transport mode irbs. */
+	if (scsw_is_tm(&irb->scsw)) {
+		memcpy(&cdev->private->irb, irb, sizeof(struct irb));
+		return;
+	}
+	/*
+	 * Don't accumulate unsolicited interrupts.
+	 */
+	if (!scsw_is_solicited(&irb->scsw))
+		return;
+
+	cdev_irb = &cdev->private->irb;
+
+	/*
+	 * If the clear function had been performed, all formerly pending
+	 * status at the subchannel has been cleared and we must not pass
+	 * intermediate accumulated status to the device driver.
+	 */
+	if (irb->scsw.cmd.fctl & SCSW_FCTL_CLEAR_FUNC)
+		memset(&cdev->private->irb, 0, sizeof(struct irb));
+
+	/* Copy bits which are valid only for the start function. */
+	if (irb->scsw.cmd.fctl & SCSW_FCTL_START_FUNC) {
+		/* Copy key. */
+		cdev_irb->scsw.cmd.key = irb->scsw.cmd.key;
+		/* Copy suspend control bit. */
+		cdev_irb->scsw.cmd.sctl = irb->scsw.cmd.sctl;
+		/* Accumulate deferred condition code. */
+		cdev_irb->scsw.cmd.cc |= irb->scsw.cmd.cc;
+		/* Copy ccw format bit. */
+		cdev_irb->scsw.cmd.fmt = irb->scsw.cmd.fmt;
+		/* Copy prefetch bit. */
+		cdev_irb->scsw.cmd.pfch = irb->scsw.cmd.pfch;
+		/* Copy initial-status-interruption-control. */
+		cdev_irb->scsw.cmd.isic = irb->scsw.cmd.isic;
+		/* Copy address limit checking control. */
+		cdev_irb->scsw.cmd.alcc = irb->scsw.cmd.alcc;
+		/* Copy suppress suspend bit. */
+		cdev_irb->scsw.cmd.ssi = irb->scsw.cmd.ssi;
+	}
+
+	/* Take care of the extended control bit and extended control word. */
+	ccw_device_accumulate_ecw(cdev, irb);
+	    
+	/* Accumulate function control. */
+	cdev_irb->scsw.cmd.fctl |= irb->scsw.cmd.fctl;
+	/* Copy activity control. */
+	cdev_irb->scsw.cmd.actl = irb->scsw.cmd.actl;
+	/* Accumulate status control. */
+	cdev_irb->scsw.cmd.stctl |= irb->scsw.cmd.stctl;
+	/*
+	 * Copy ccw address if it is valid. This is a bit simplified
+	 * but should be close enough for all practical purposes.
+	 */
+	if ((irb->scsw.cmd.stctl & SCSW_STCTL_PRIM_STATUS) ||
+	    ((irb->scsw.cmd.stctl ==
+	      (SCSW_STCTL_INTER_STATUS|SCSW_STCTL_STATUS_PEND)) &&
+	     (irb->scsw.cmd.actl & SCSW_ACTL_DEVACT) &&
+	     (irb->scsw.cmd.actl & SCSW_ACTL_SCHACT)) ||
+	    (irb->scsw.cmd.actl & SCSW_ACTL_SUSPENDED))
+		cdev_irb->scsw.cmd.cpa = irb->scsw.cmd.cpa;
+	/* Accumulate device status, but not the device busy flag. */
+	cdev_irb->scsw.cmd.dstat &= ~DEV_STAT_BUSY;
+	/* dstat is not always valid. */
+	if (irb->scsw.cmd.stctl &
+	    (SCSW_STCTL_PRIM_STATUS | SCSW_STCTL_SEC_STATUS
+	     | SCSW_STCTL_INTER_STATUS | SCSW_STCTL_ALERT_STATUS))
+		cdev_irb->scsw.cmd.dstat |= irb->scsw.cmd.dstat;
+	/* Accumulate subchannel status. */
+	cdev_irb->scsw.cmd.cstat |= irb->scsw.cmd.cstat;
+	/* Copy residual count if it is valid. */
+	if ((irb->scsw.cmd.stctl & SCSW_STCTL_PRIM_STATUS) &&
+	    (irb->scsw.cmd.cstat & ~(SCHN_STAT_PCI | SCHN_STAT_INCORR_LEN))
+	     == 0)
+		cdev_irb->scsw.cmd.count = irb->scsw.cmd.count;
+
+	/* Take care of bits in the extended status word. */
+	ccw_device_accumulate_esw(cdev, irb);
+
+	/*
+	 * Check whether we must issue a SENSE CCW ourselves if there is no
+	 * concurrent sense facility installed for the subchannel.
+	 * No sense is required if no delayed sense is pending
+	 * and we did not get a unit check without sense information.
+	 *
+	 * Note: We should check for ioinfo[irq]->flags.consns but VM
+	 *	 violates the ESA/390 architecture and doesn't present an
+	 *	 operand exception for virtual devices without concurrent
+	 *	 sense facility available/supported when enabling the
+	 *	 concurrent sense facility.
+	 */
+	if ((cdev_irb->scsw.cmd.dstat & DEV_STAT_UNIT_CHECK) &&
+	    !(cdev_irb->esw.esw0.erw.cons))
+		cdev->private->flags.dosense = 1;
+}
+
+/*
+ * Do a basic sense.
+ */
+int
+ccw_device_do_sense(struct ccw_device *cdev, struct irb *irb)
+{
+	struct subchannel *sch;
+	struct ccw1 *sense_ccw;
+	int rc;
+
+	sch = to_subchannel(cdev->dev.parent);
+
+	/* A sense is required, can we do it now ? */
+	if (scsw_actl(&irb->scsw) & (SCSW_ACTL_DEVACT | SCSW_ACTL_SCHACT))
+		/*
+		 * we received an Unit Check but we have no final
+		 *  status yet, therefore we must delay the SENSE
+		 *  processing. We must not report this intermediate
+		 *  status to the device interrupt handler.
+		 */
+		return -EBUSY;
+
+	/*
+	 * We have ending status but no sense information. Do a basic sense.
+	 */
+	sense_ccw = &to_io_private(sch)->sense_ccw;
+	sense_ccw->cmd_code = CCW_CMD_BASIC_SENSE;
+	sense_ccw->cda = (__u32) __pa(cdev->private->irb.ecw);
+	sense_ccw->count = SENSE_MAX_COUNT;
+	sense_ccw->flags = CCW_FLAG_SLI;
+
+	/* Reset internal retry indication. */
+	cdev->private->flags.intretry = 0;
+
+	rc = cio_start(sch, sense_ccw, 0xff);
+	if (rc == -ENODEV || rc == -EACCES)
+		dev_fsm_event(cdev, DEV_EVENT_VERIFY);
+	return rc;
+}
+
+/*
+ * Add information from basic sense to devstat.
+ */
+void
+ccw_device_accumulate_basic_sense(struct ccw_device *cdev, struct irb *irb)
+{
+	/*
+	 * Check if the status pending bit is set in stctl.
+	 * If not, the remaining bit have no meaning and we must ignore them.
+	 * The esw is not meaningful as well...
+	 */
+	if (!(scsw_stctl(&irb->scsw) & SCSW_STCTL_STATUS_PEND))
+		return;
+
+	/* Check for channel checks and interface control checks. */
+	ccw_device_msg_control_check(cdev, irb);
+
+	/* Check for path not operational. */
+	if (scsw_is_valid_pno(&irb->scsw) && scsw_pno(&irb->scsw))
+		ccw_device_path_notoper(cdev);
+
+	if (!(irb->scsw.cmd.dstat & DEV_STAT_UNIT_CHECK) &&
+	    (irb->scsw.cmd.dstat & DEV_STAT_CHN_END)) {
+		cdev->private->irb.esw.esw0.erw.cons = 1;
+		cdev->private->flags.dosense = 0;
+	}
+	/* Check if path verification is required. */
+	if (ccw_device_accumulate_esw_valid(irb) &&
+	    irb->esw.esw0.erw.pvrf)
+		cdev->private->flags.doverify = 1;
+}
+
+/*
+ * This function accumulates the status into the private devstat and
+ * starts a basic sense if one is needed.
+ */
+int
+ccw_device_accumulate_and_sense(struct ccw_device *cdev, struct irb *irb)
+{
+	ccw_device_accumulate_irb(cdev, irb);
+	if ((irb->scsw.cmd.actl  & (SCSW_ACTL_DEVACT | SCSW_ACTL_SCHACT)) != 0)
+		return -EBUSY;
+	/* Check for basic sense. */
+	if (cdev->private->flags.dosense &&
+	    !(irb->scsw.cmd.dstat & DEV_STAT_UNIT_CHECK)) {
+		cdev->private->irb.esw.esw0.erw.cons = 1;
+		cdev->private->flags.dosense = 0;
+		return 0;
+	}
+	if (cdev->private->flags.dosense) {
+		ccw_device_do_sense(cdev, irb);
+		return -EBUSY;
+	}
+	return 0;
+}
+

kernel/drivers/s390/cio/fcx.c

@@ -0,0 +1,350 @@
+/*
+ *  Functions for assembling fcx enabled I/O control blocks.
+ *
+ *    Copyright IBM Corp. 2008
+ *    Author(s): Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/err.h>
+#include <linux/module.h>
+#include <asm/fcx.h>
+#include "cio.h"
+
+/**
+ * tcw_get_intrg - return pointer to associated interrogate tcw
+ * @tcw: pointer to the original tcw
+ *
+ * Return a pointer to the interrogate tcw associated with the specified tcw
+ * or %NULL if there is no associated interrogate tcw.
+ */
+struct tcw *tcw_get_intrg(struct tcw *tcw)
+{
+	return (struct tcw *) ((addr_t) tcw->intrg);
+}
+EXPORT_SYMBOL(tcw_get_intrg);
+
+/**
+ * tcw_get_data - return pointer to input/output data associated with tcw
+ * @tcw: pointer to the tcw
+ *
+ * Return the input or output data address specified in the tcw depending
+ * on whether the r-bit or the w-bit is set. If neither bit is set, return
+ * %NULL.
+ */
+void *tcw_get_data(struct tcw *tcw)
+{
+	if (tcw->r)
+		return (void *) ((addr_t) tcw->input);
+	if (tcw->w)
+		return (void *) ((addr_t) tcw->output);
+	return NULL;
+}
+EXPORT_SYMBOL(tcw_get_data);
+
+/**
+ * tcw_get_tccb - return pointer to tccb associated with tcw
+ * @tcw: pointer to the tcw
+ *
+ * Return pointer to the tccb associated with this tcw.
+ */
+struct tccb *tcw_get_tccb(struct tcw *tcw)
+{
+	return (struct tccb *) ((addr_t) tcw->tccb);
+}
+EXPORT_SYMBOL(tcw_get_tccb);
+
+/**
+ * tcw_get_tsb - return pointer to tsb associated with tcw
+ * @tcw: pointer to the tcw
+ *
+ * Return pointer to the tsb associated with this tcw.
+ */
+struct tsb *tcw_get_tsb(struct tcw *tcw)
+{
+	return (struct tsb *) ((addr_t) tcw->tsb);
+}
+EXPORT_SYMBOL(tcw_get_tsb);
+
+/**
+ * tcw_init - initialize tcw data structure
+ * @tcw: pointer to the tcw to be initialized
+ * @r: initial value of the r-bit
+ * @w: initial value of the w-bit
+ *
+ * Initialize all fields of the specified tcw data structure with zero and
+ * fill in the format, flags, r and w fields.
+ */
+void tcw_init(struct tcw *tcw, int r, int w)
+{
+	memset(tcw, 0, sizeof(struct tcw));
+	tcw->format = TCW_FORMAT_DEFAULT;
+	tcw->flags = TCW_FLAGS_TIDAW_FORMAT(TCW_TIDAW_FORMAT_DEFAULT);
+	if (r)
+		tcw->r = 1;
+	if (w)
+		tcw->w = 1;
+}
+EXPORT_SYMBOL(tcw_init);
+
+static inline size_t tca_size(struct tccb *tccb)
+{
+	return tccb->tcah.tcal - 12;
+}
+
+static u32 calc_dcw_count(struct tccb *tccb)
+{
+	int offset;
+	struct dcw *dcw;
+	u32 count = 0;
+	size_t size;
+
+	size = tca_size(tccb);
+	for (offset = 0; offset < size;) {
+		dcw = (struct dcw *) &tccb->tca[offset];
+		count += dcw->count;
+		if (!(dcw->flags & DCW_FLAGS_CC))
+			break;
+		offset += sizeof(struct dcw) + ALIGN((int) dcw->cd_count, 4);
+	}
+	return count;
+}
+
+static u32 calc_cbc_size(struct tidaw *tidaw, int num)
+{
+	int i;
+	u32 cbc_data;
+	u32 cbc_count = 0;
+	u64 data_count = 0;
+
+	for (i = 0; i < num; i++) {
+		if (tidaw[i].flags & TIDAW_FLAGS_LAST)
+			break;
+		/* TODO: find out if padding applies to total of data
+		 * transferred or data transferred by this tidaw. Assumption:
+		 * applies to total. */
+		data_count += tidaw[i].count;
+		if (tidaw[i].flags & TIDAW_FLAGS_INSERT_CBC) {
+			cbc_data = 4 + ALIGN(data_count, 4) - data_count;
+			cbc_count += cbc_data;
+			data_count += cbc_data;
+		}
+	}
+	return cbc_count;
+}
+
+/**
+ * tcw_finalize - finalize tcw length fields and tidaw list
+ * @tcw: pointer to the tcw
+ * @num_tidaws: the number of tidaws used to address input/output data or zero
+ * if no tida is used
+ *
+ * Calculate the input-/output-count and tccbl field in the tcw, add a
+ * tcat the tccb and terminate the data tidaw list if used.
+ *
+ * Note: in case input- or output-tida is used, the tidaw-list must be stored
+ * in contiguous storage (no ttic). The tcal field in the tccb must be
+ * up-to-date.
+ */
+void tcw_finalize(struct tcw *tcw, int num_tidaws)
+{
+	struct tidaw *tidaw;
+	struct tccb *tccb;
+	struct tccb_tcat *tcat;
+	u32 count;
+
+	/* Terminate tidaw list. */
+	tidaw = tcw_get_data(tcw);
+	if (num_tidaws > 0)
+		tidaw[num_tidaws - 1].flags |= TIDAW_FLAGS_LAST;
+	/* Add tcat to tccb. */
+	tccb = tcw_get_tccb(tcw);
+	tcat = (struct tccb_tcat *) &tccb->tca[tca_size(tccb)];
+	memset(tcat, 0, sizeof(tcat));
+	/* Calculate tcw input/output count and tcat transport count. */
+	count = calc_dcw_count(tccb);
+	if (tcw->w && (tcw->flags & TCW_FLAGS_OUTPUT_TIDA))
+		count += calc_cbc_size(tidaw, num_tidaws);
+	if (tcw->r)
+		tcw->input_count = count;
+	else if (tcw->w)
+		tcw->output_count = count;
+	tcat->count = ALIGN(count, 4) + 4;
+	/* Calculate tccbl. */
+	tcw->tccbl = (sizeof(struct tccb) + tca_size(tccb) +
+		      sizeof(struct tccb_tcat) - 20) >> 2;
+}
+EXPORT_SYMBOL(tcw_finalize);
+
+/**
+ * tcw_set_intrg - set the interrogate tcw address of a tcw
+ * @tcw: the tcw address
+ * @intrg_tcw: the address of the interrogate tcw
+ *
+ * Set the address of the interrogate tcw in the specified tcw.
+ */
+void tcw_set_intrg(struct tcw *tcw, struct tcw *intrg_tcw)
+{
+	tcw->intrg = (u32) ((addr_t) intrg_tcw);
+}
+EXPORT_SYMBOL(tcw_set_intrg);
+
+/**
+ * tcw_set_data - set data address and tida flag of a tcw
+ * @tcw: the tcw address
+ * @data: the data address
+ * @use_tidal: zero of the data address specifies a contiguous block of data,
+ * non-zero if it specifies a list if tidaws.
+ *
+ * Set the input/output data address of a tcw (depending on the value of the
+ * r-flag and w-flag). If @use_tidal is non-zero, the corresponding tida flag
+ * is set as well.
+ */
+void tcw_set_data(struct tcw *tcw, void *data, int use_tidal)
+{
+	if (tcw->r) {
+		tcw->input = (u64) ((addr_t) data);
+		if (use_tidal)
+			tcw->flags |= TCW_FLAGS_INPUT_TIDA;
+	} else if (tcw->w) {
+		tcw->output = (u64) ((addr_t) data);
+		if (use_tidal)
+			tcw->flags |= TCW_FLAGS_OUTPUT_TIDA;
+	}
+}
+EXPORT_SYMBOL(tcw_set_data);
+
+/**
+ * tcw_set_tccb - set tccb address of a tcw
+ * @tcw: the tcw address
+ * @tccb: the tccb address
+ *
+ * Set the address of the tccb in the specified tcw.
+ */
+void tcw_set_tccb(struct tcw *tcw, struct tccb *tccb)
+{
+	tcw->tccb = (u64) ((addr_t) tccb);
+}
+EXPORT_SYMBOL(tcw_set_tccb);
+
+/**
+ * tcw_set_tsb - set tsb address of a tcw
+ * @tcw: the tcw address
+ * @tsb: the tsb address
+ *
+ * Set the address of the tsb in the specified tcw.
+ */
+void tcw_set_tsb(struct tcw *tcw, struct tsb *tsb)
+{
+	tcw->tsb = (u64) ((addr_t) tsb);
+}
+EXPORT_SYMBOL(tcw_set_tsb);
+
+/**
+ * tccb_init - initialize tccb
+ * @tccb: the tccb address
+ * @size: the maximum size of the tccb
+ * @sac: the service-action-code to be user
+ *
+ * Initialize the header of the specified tccb by resetting all values to zero
+ * and filling in defaults for format, sac and initial tcal fields.
+ */
+void tccb_init(struct tccb *tccb, size_t size, u32 sac)
+{
+	memset(tccb, 0, size);
+	tccb->tcah.format = TCCB_FORMAT_DEFAULT;
+	tccb->tcah.sac = sac;
+	tccb->tcah.tcal = 12;
+}
+EXPORT_SYMBOL(tccb_init);
+
+/**
+ * tsb_init - initialize tsb
+ * @tsb: the tsb address
+ *
+ * Initialize the specified tsb by resetting all values to zero.
+ */
+void tsb_init(struct tsb *tsb)
+{
+	memset(tsb, 0, sizeof(tsb));
+}
+EXPORT_SYMBOL(tsb_init);
+
+/**
+ * tccb_add_dcw - add a dcw to the tccb
+ * @tccb: the tccb address
+ * @tccb_size: the maximum tccb size
+ * @cmd: the dcw command
+ * @flags: flags for the dcw
+ * @cd: pointer to control data for this dcw or NULL if none is required
+ * @cd_count: number of control data bytes for this dcw
+ * @count: number of data bytes for this dcw
+ *
+ * Add a new dcw to the specified tccb by writing the dcw information specified
+ * by @cmd, @flags, @cd, @cd_count and @count to the tca of the tccb. Return
+ * a pointer to the newly added dcw on success or -%ENOSPC if the new dcw
+ * would exceed the available space as defined by @tccb_size.
+ *
+ * Note: the tcal field of the tccb header will be updates to reflect added
+ * content.
+ */
+struct dcw *tccb_add_dcw(struct tccb *tccb, size_t tccb_size, u8 cmd, u8 flags,
+			 void *cd, u8 cd_count, u32 count)
+{
+	struct dcw *dcw;
+	int size;
+	int tca_offset;
+
+	/* Check for space. */
+	tca_offset = tca_size(tccb);
+	size = ALIGN(sizeof(struct dcw) + cd_count, 4);
+	if (sizeof(struct tccb_tcah) + tca_offset + size +
+	    sizeof(struct tccb_tcat) > tccb_size)
+		return ERR_PTR(-ENOSPC);
+	/* Add dcw to tca. */
+	dcw = (struct dcw *) &tccb->tca[tca_offset];
+	memset(dcw, 0, size);
+	dcw->cmd = cmd;
+	dcw->flags = flags;
+	dcw->count = count;
+	dcw->cd_count = cd_count;
+	if (cd)
+		memcpy(&dcw->cd[0], cd, cd_count);
+	tccb->tcah.tcal += size;
+	return dcw;
+}
+EXPORT_SYMBOL(tccb_add_dcw);
+
+/**
+ * tcw_add_tidaw - add a tidaw to a tcw
+ * @tcw: the tcw address
+ * @num_tidaws: the current number of tidaws
+ * @flags: flags for the new tidaw
+ * @addr: address value for the new tidaw
+ * @count: count value for the new tidaw
+ *
+ * Add a new tidaw to the input/output data tidaw-list of the specified tcw
+ * (depending on the value of the r-flag and w-flag) and return a pointer to
+ * the new tidaw.
+ *
+ * Note: the tidaw-list is assumed to be contiguous with no ttics. The caller
+ * must ensure that there is enough space for the new tidaw. The last-tidaw
+ * flag for the last tidaw in the list will be set by tcw_finalize.
+ */
+struct tidaw *tcw_add_tidaw(struct tcw *tcw, int num_tidaws, u8 flags,
+			    void *addr, u32 count)
+{
+	struct tidaw *tidaw;
+
+	/* Add tidaw to tidaw-list. */
+	tidaw = ((struct tidaw *) tcw_get_data(tcw)) + num_tidaws;
+	memset(tidaw, 0, sizeof(struct tidaw));
+	tidaw->flags = flags;
+	tidaw->count = count;
+	tidaw->addr = (u64) ((addr_t) addr);
+	return tidaw;
+}
+EXPORT_SYMBOL(tcw_add_tidaw);

kernel/drivers/s390/cio/idset.c

@@ -0,0 +1,132 @@
+/*
+ *  drivers/s390/cio/idset.c
+ *
+ *    Copyright IBM Corp. 2007
+ *    Author(s): Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
+ */
+
+#include <linux/vmalloc.h>
+#include <linux/bitops.h>
+#include "idset.h"
+#include "css.h"
+
+struct idset {
+	int num_ssid;
+	int num_id;
+	unsigned long bitmap[0];
+};
+
+static inline unsigned long bitmap_size(int num_ssid, int num_id)
+{
+	return __BITOPS_WORDS(num_ssid * num_id) * sizeof(unsigned long);
+}
+
+static struct idset *idset_new(int num_ssid, int num_id)
+{
+	struct idset *set;
+
+	set = vmalloc(sizeof(struct idset) + bitmap_size(num_ssid, num_id));
+	if (set) {
+		set->num_ssid = num_ssid;
+		set->num_id = num_id;
+		memset(set->bitmap, 0, bitmap_size(num_ssid, num_id));
+	}
+	return set;
+}
+
+void idset_free(struct idset *set)
+{
+	vfree(set);
+}
+
+void idset_clear(struct idset *set)
+{
+	memset(set->bitmap, 0, bitmap_size(set->num_ssid, set->num_id));
+}
+
+void idset_fill(struct idset *set)
+{
+	memset(set->bitmap, 0xff, bitmap_size(set->num_ssid, set->num_id));
+}
+
+static inline void idset_add(struct idset *set, int ssid, int id)
+{
+	set_bit(ssid * set->num_id + id, set->bitmap);
+}
+
+static inline void idset_del(struct idset *set, int ssid, int id)
+{
+	clear_bit(ssid * set->num_id + id, set->bitmap);
+}
+
+static inline int idset_contains(struct idset *set, int ssid, int id)
+{
+	return test_bit(ssid * set->num_id + id, set->bitmap);
+}
+
+static inline int idset_get_first(struct idset *set, int *ssid, int *id)
+{
+	int bitnum;
+
+	bitnum = find_first_bit(set->bitmap, set->num_ssid * set->num_id);
+	if (bitnum >= set->num_ssid * set->num_id)
+		return 0;
+	*ssid = bitnum / set->num_id;
+	*id = bitnum % set->num_id;
+	return 1;
+}
+
+struct idset *idset_sch_new(void)
+{
+	return idset_new(max_ssid + 1, __MAX_SUBCHANNEL + 1);
+}
+
+void idset_sch_add(struct idset *set, struct subchannel_id schid)
+{
+	idset_add(set, schid.ssid, schid.sch_no);
+}
+
+void idset_sch_del(struct idset *set, struct subchannel_id schid)
+{
+	idset_del(set, schid.ssid, schid.sch_no);
+}
+
+int idset_sch_contains(struct idset *set, struct subchannel_id schid)
+{
+	return idset_contains(set, schid.ssid, schid.sch_no);
+}
+
+int idset_sch_get_first(struct idset *set, struct subchannel_id *schid)
+{
+	int ssid = 0;
+	int id = 0;
+	int rc;
+
+	rc = idset_get_first(set, &ssid, &id);
+	if (rc) {
+		init_subchannel_id(schid);
+		schid->ssid = ssid;
+		schid->sch_no = id;
+	}
+	return rc;
+}
+
+int idset_is_empty(struct idset *set)
+{
+	int bitnum;
+
+	bitnum = find_first_bit(set->bitmap, set->num_ssid * set->num_id);
+	if (bitnum >= set->num_ssid * set->num_id)
+		return 1;
+	return 0;
+}
+
+void idset_add_set(struct idset *to, struct idset *from)
+{
+	unsigned long i, len;
+
+	len = min(__BITOPS_WORDS(to->num_ssid * to->num_id),
+		  __BITOPS_WORDS(from->num_ssid * from->num_id));
+	for (i = 0; i < len ; i++)
+		to->bitmap[i] |= from->bitmap[i];
+}