7 files changed, 2259 insertions, 0 deletions

diff --git a/arch/s390/pci/Makefile b/arch/s390/pci/Makefile
new file mode 100644
index 000000000000..ab0827b6bc4b
--- /dev/null
+++ b/arch/s390/pci/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for the s390 PCI subsystem.
+#
+
+obj-$(CONFIG_PCI)	+= pci.o pci_dma.o pci_clp.o pci_msi.o \
+			   pci_sysfs.o pci_event.o
diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
new file mode 100644
index 000000000000..7ed38e5e3028
--- /dev/null
+++ b/arch/s390/pci/pci.c
@@ -0,0 +1,1103 @@
+/*
+ * Copyright IBM Corp. 2012
+ *
+ * Author(s):
+ *   Jan Glauber <jang@linux.vnet.ibm.com>
+ *
+ * The System z PCI code is a rewrite from a prototype by
+ * the following people (Kudoz!):
+ *   Alexander Schmidt
+ *   Christoph Raisch
+ *   Hannes Hering
+ *   Hoang-Nam Nguyen
+ *   Jan-Bernd Themann
+ *   Stefan Roscher
+ *   Thomas Klein
+ */
+
+#define COMPONENT "zPCI"
+#define pr_fmt(fmt) COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/err.h>
+#include <linux/export.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+#include <linux/kernel_stat.h>
+#include <linux/seq_file.h>
+#include <linux/pci.h>
+#include <linux/msi.h>
+
+#include <asm/isc.h>
+#include <asm/airq.h>
+#include <asm/facility.h>
+#include <asm/pci_insn.h>
+#include <asm/pci_clp.h>
+#include <asm/pci_dma.h>
+
+#define DEBUG				/* enable pr_debug */
+
+#define	SIC_IRQ_MODE_ALL		0
+#define	SIC_IRQ_MODE_SINGLE		1
+
+#define ZPCI_NR_DMA_SPACES		1
+#define ZPCI_MSI_VEC_BITS		6
+#define ZPCI_NR_DEVICES			CONFIG_PCI_NR_FUNCTIONS
+
+/* list of all detected zpci devices */
+LIST_HEAD(zpci_list);
+EXPORT_SYMBOL_GPL(zpci_list);
+DEFINE_MUTEX(zpci_list_lock);
+EXPORT_SYMBOL_GPL(zpci_list_lock);
+
+struct pci_hp_callback_ops hotplug_ops;
+EXPORT_SYMBOL_GPL(hotplug_ops);
+
+static DECLARE_BITMAP(zpci_domain, ZPCI_NR_DEVICES);
+static DEFINE_SPINLOCK(zpci_domain_lock);
+
+struct callback {
+	irq_handler_t	handler;
+	void		*data;
+};
+
+struct zdev_irq_map {
+	unsigned long	aibv;		/* AI bit vector */
+	int		msi_vecs;	/* consecutive MSI-vectors used */
+	int		__unused;
+	struct callback	cb[ZPCI_NR_MSI_VECS]; /* callback handler array */
+	spinlock_t	lock;		/* protect callbacks against de-reg */
+};
+
+struct intr_bucket {
+	/* amap of adapters, one bit per dev, corresponds to one irq nr */
+	unsigned long	*alloc;
+	/* AI summary bit, global page for all devices */
+	unsigned long	*aisb;
+	/* pointer to aibv and callback data in zdev */
+	struct zdev_irq_map *imap[ZPCI_NR_DEVICES];
+	/* protects the whole bucket struct */
+	spinlock_t	lock;
+};
+
+static struct intr_bucket *bucket;
+
+/* Adapter local summary indicator */
+static u8 *zpci_irq_si;
+
+static atomic_t irq_retries = ATOMIC_INIT(0);
+
+/* I/O Map */
+static DEFINE_SPINLOCK(zpci_iomap_lock);
+static DECLARE_BITMAP(zpci_iomap, ZPCI_IOMAP_MAX_ENTRIES);
+struct zpci_iomap_entry *zpci_iomap_start;
+EXPORT_SYMBOL_GPL(zpci_iomap_start);
+
+/* highest irq summary bit */
+static int __read_mostly aisb_max;
+
+static struct kmem_cache *zdev_irq_cache;
+
+static inline int irq_to_msi_nr(unsigned int irq)
+{
+	return irq & ZPCI_MSI_MASK;
+}
+
+static inline int irq_to_dev_nr(unsigned int irq)
+{
+	return irq >> ZPCI_MSI_VEC_BITS;
+}
+
+static inline struct zdev_irq_map *get_imap(unsigned int irq)
+{
+	return bucket->imap[irq_to_dev_nr(irq)];
+}
+
+struct zpci_dev *get_zdev(struct pci_dev *pdev)
+{
+	return (struct zpci_dev *) pdev->sysdata;
+}
+
+struct zpci_dev *get_zdev_by_fid(u32 fid)
+{
+	struct zpci_dev *tmp, *zdev = NULL;
+
+	mutex_lock(&zpci_list_lock);
+	list_for_each_entry(tmp, &zpci_list, entry) {
+		if (tmp->fid == fid) {
+			zdev = tmp;
+			break;
+		}
+	}
+	mutex_unlock(&zpci_list_lock);
+	return zdev;
+}
+
+bool zpci_fid_present(u32 fid)
+{
+	return (get_zdev_by_fid(fid) != NULL) ? true : false;
+}
+
+static struct zpci_dev *get_zdev_by_bus(struct pci_bus *bus)
+{
+	return (bus && bus->sysdata) ? (struct zpci_dev *) bus->sysdata : NULL;
+}
+
+int pci_domain_nr(struct pci_bus *bus)
+{
+	return ((struct zpci_dev *) bus->sysdata)->domain;
+}
+EXPORT_SYMBOL_GPL(pci_domain_nr);
+
+int pci_proc_domain(struct pci_bus *bus)
+{
+	return pci_domain_nr(bus);
+}
+EXPORT_SYMBOL_GPL(pci_proc_domain);
+
+/* Store PCI function information block */
+static int zpci_store_fib(struct zpci_dev *zdev, u8 *fc)
+{
+	struct zpci_fib *fib;
+	u8 status, cc;
+
+	fib = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!fib)
+		return -ENOMEM;
+
+	do {
+		cc = __stpcifc(zdev->fh, 0, fib, &status);
+		if (cc == 2) {
+			msleep(ZPCI_INSN_BUSY_DELAY);
+			memset(fib, 0, PAGE_SIZE);
+		}
+	} while (cc == 2);
+
+	if (cc)
+		pr_err_once("%s: cc: %u  status: %u\n",
+			    __func__, cc, status);
+
+	/* Return PCI function controls */
+	*fc = fib->fc;
+
+	free_page((unsigned long) fib);
+	return (cc) ? -EIO : 0;
+}
+
+/* Modify PCI: Register adapter interruptions */
+static int zpci_register_airq(struct zpci_dev *zdev, unsigned int aisb,
+			      u64 aibv)
+{
+	u64 req = ZPCI_CREATE_REQ(zdev->fh, 0, ZPCI_MOD_FC_REG_INT);
+	struct zpci_fib *fib;
+	int rc;
+
+	fib = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!fib)
+		return -ENOMEM;
+
+	fib->isc = PCI_ISC;
+	fib->noi = zdev->irq_map->msi_vecs;
+	fib->sum = 1;		/* enable summary notifications */
+	fib->aibv = aibv;
+	fib->aibvo = 0;		/* every function has its own page */
+	fib->aisb = (u64) bucket->aisb + aisb / 8;
+	fib->aisbo = aisb & ZPCI_MSI_MASK;
+
+	rc = mpcifc_instr(req, fib);
+	pr_debug("%s mpcifc returned noi: %d\n", __func__, fib->noi);
+
+	free_page((unsigned long) fib);
+	return rc;
+}
+
+struct mod_pci_args {
+	u64 base;
+	u64 limit;
+	u64 iota;
+};
+
+static int mod_pci(struct zpci_dev *zdev, int fn, u8 dmaas, struct mod_pci_args *args)
+{
+	u64 req = ZPCI_CREATE_REQ(zdev->fh, dmaas, fn);
+	struct zpci_fib *fib;
+	int rc;
+
+	/* The FIB must be available even if it's not used */
+	fib = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!fib)
+		return -ENOMEM;
+
+	fib->pba = args->base;
+	fib->pal = args->limit;
+	fib->iota = args->iota;
+
+	rc = mpcifc_instr(req, fib);
+	free_page((unsigned long) fib);
+	return rc;
+}
+
+/* Modify PCI: Register I/O address translation parameters */
+int zpci_register_ioat(struct zpci_dev *zdev, u8 dmaas,
+		       u64 base, u64 limit, u64 iota)
+{
+	struct mod_pci_args args = { base, limit, iota };
+
+	WARN_ON_ONCE(iota & 0x3fff);
+	args.iota |= ZPCI_IOTA_RTTO_FLAG;
+	return mod_pci(zdev, ZPCI_MOD_FC_REG_IOAT, dmaas, &args);
+}
+
+/* Modify PCI: Unregister I/O address translation parameters */
+int zpci_unregister_ioat(struct zpci_dev *zdev, u8 dmaas)
+{
+	struct mod_pci_args args = { 0, 0, 0 };
+
+	return mod_pci(zdev, ZPCI_MOD_FC_DEREG_IOAT, dmaas, &args);
+}
+
+/* Modify PCI: Unregister adapter interruptions */
+static int zpci_unregister_airq(struct zpci_dev *zdev)
+{
+	struct mod_pci_args args = { 0, 0, 0 };
+
+	return mod_pci(zdev, ZPCI_MOD_FC_DEREG_INT, 0, &args);
+}
+
+#define ZPCI_PCIAS_CFGSPC	15
+
+static int zpci_cfg_load(struct zpci_dev *zdev, int offset, u32 *val, u8 len)
+{
+	u64 req = ZPCI_CREATE_REQ(zdev->fh, ZPCI_PCIAS_CFGSPC, len);
+	u64 data;
+	int rc;
+
+	rc = pcilg_instr(&data, req, offset);
+	data = data << ((8 - len) * 8);
+	data = le64_to_cpu(data);
+	if (!rc)
+		*val = (u32) data;
+	else
+		*val = 0xffffffff;
+	return rc;
+}
+
+static int zpci_cfg_store(struct zpci_dev *zdev, int offset, u32 val, u8 len)
+{
+	u64 req = ZPCI_CREATE_REQ(zdev->fh, ZPCI_PCIAS_CFGSPC, len);
+	u64 data = val;
+	int rc;
+
+	data = cpu_to_le64(data);
+	data = data >> ((8 - len) * 8);
+	rc = pcistg_instr(data, req, offset);
+	return rc;
+}
+
+void synchronize_irq(unsigned int irq)
+{
+	/*
+	 * Not needed, the handler is protected by a lock and IRQs that occur
+	 * after the handler is deleted are just NOPs.
+	 */
+}
+EXPORT_SYMBOL_GPL(synchronize_irq);
+
+void enable_irq(unsigned int irq)
+{
+	struct msi_desc *msi = irq_get_msi_desc(irq);
+
+	zpci_msi_set_mask_bits(msi, 1, 0);
+}
+EXPORT_SYMBOL_GPL(enable_irq);
+
+void disable_irq(unsigned int irq)
+{
+	struct msi_desc *msi = irq_get_msi_desc(irq);
+
+	zpci_msi_set_mask_bits(msi, 1, 1);
+}
+EXPORT_SYMBOL_GPL(disable_irq);
+
+void disable_irq_nosync(unsigned int irq)
+{
+	disable_irq(irq);
+}
+EXPORT_SYMBOL_GPL(disable_irq_nosync);
+
+unsigned long probe_irq_on(void)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(probe_irq_on);
+
+int probe_irq_off(unsigned long val)
+{
+	return 0;
+}
+EXPORT_SYMBOL_GPL(probe_irq_off);
+
+unsigned int probe_irq_mask(unsigned long val)
+{
+	return val;
+}
+EXPORT_SYMBOL_GPL(probe_irq_mask);
+
+void __devinit pcibios_fixup_bus(struct pci_bus *bus)
+{
+}
+
+resource_size_t pcibios_align_resource(void *data, const struct resource *res,
+				       resource_size_t size,
+				       resource_size_t align)
+{
+	return 0;
+}
+
+/* combine single writes by using store-block insn */
+void __iowrite64_copy(void __iomem *to, const void *from, size_t count)
+{
+       zpci_memcpy_toio(to, from, count);
+}
+
+/* Create a virtual mapping cookie for a PCI BAR */
+void __iomem *pci_iomap(struct pci_dev *pdev, int bar, unsigned long max)
+{
+	struct zpci_dev *zdev =	get_zdev(pdev);
+	u64 addr;
+	int idx;
+
+	if ((bar & 7) != bar)
+		return NULL;
+
+	idx = zdev->bars[bar].map_idx;
+	spin_lock(&zpci_iomap_lock);
+	zpci_iomap_start[idx].fh = zdev->fh;
+	zpci_iomap_start[idx].bar = bar;
+	spin_unlock(&zpci_iomap_lock);
+
+	addr = ZPCI_IOMAP_ADDR_BASE | ((u64) idx << 48);
+	return (void __iomem *) addr;
+}
+EXPORT_SYMBOL_GPL(pci_iomap);
+
+void pci_iounmap(struct pci_dev *pdev, void __iomem *addr)
+{
+	unsigned int idx;
+
+	idx = (((__force u64) addr) & ~ZPCI_IOMAP_ADDR_BASE) >> 48;
+	spin_lock(&zpci_iomap_lock);
+	zpci_iomap_start[idx].fh = 0;
+	zpci_iomap_start[idx].bar = 0;
+	spin_unlock(&zpci_iomap_lock);
+}
+EXPORT_SYMBOL_GPL(pci_iounmap);
+
+static int pci_read(struct pci_bus *bus, unsigned int devfn, int where,
+		    int size, u32 *val)
+{
+	struct zpci_dev *zdev = get_zdev_by_bus(bus);
+
+	if (!zdev || devfn != ZPCI_DEVFN)
+		return 0;
+	return zpci_cfg_load(zdev, where, val, size);
+}
+
+static int pci_write(struct pci_bus *bus, unsigned int devfn, int where,
+		     int size, u32 val)
+{
+	struct zpci_dev *zdev = get_zdev_by_bus(bus);
+
+	if (!zdev || devfn != ZPCI_DEVFN)
+		return 0;
+	return zpci_cfg_store(zdev, where, val, size);
+}
+
+static struct pci_ops pci_root_ops = {
+	.read = pci_read,
+	.write = pci_write,
+};
+
+/* store the last handled bit to implement fair scheduling of devices */
+static DEFINE_PER_CPU(unsigned long, next_sbit);
+
+static void zpci_irq_handler(void *dont, void *need)
+{
+	unsigned long sbit, mbit, last = 0, start = __get_cpu_var(next_sbit);
+	int rescan = 0, max = aisb_max;
+	struct zdev_irq_map *imap;
+
+	kstat_cpu(smp_processor_id()).irqs[IOINT_PCI]++;
+	sbit = start;
+
+scan:
+	/* find summary_bit */
+	for_each_set_bit_left_cont(sbit, bucket->aisb, max) {
+		clear_bit(63 - (sbit & 63), bucket->aisb + (sbit >> 6));
+		last = sbit;
+
+		/* find vector bit */
+		imap = bucket->imap[sbit];
+		for_each_set_bit_left(mbit, &imap->aibv, imap->msi_vecs) {
+			kstat_cpu(smp_processor_id()).irqs[IOINT_MSI]++;
+			clear_bit(63 - mbit, &imap->aibv);
+
+			spin_lock(&imap->lock);
+			if (imap->cb[mbit].handler)
+				imap->cb[mbit].handler(mbit,
+					imap->cb[mbit].data);
+			spin_unlock(&imap->lock);
+		}
+	}
+
+	if (rescan)
+		goto out;
+
+	/* scan the skipped bits */
+	if (start > 0) {
+		sbit = 0;
+		max = start;
+		start = 0;
+		goto scan;
+	}
+
+	/* enable interrupts again */
+	sic_instr(SIC_IRQ_MODE_SINGLE, NULL, PCI_ISC);
+
+	/* check again to not lose initiative */
+	rmb();
+	max = aisb_max;
+	sbit = find_first_bit_left(bucket->aisb, max);
+	if (sbit != max) {
+		atomic_inc(&irq_retries);
+		rescan++;
+		goto scan;
+	}
+out:
+	/* store next device bit to scan */
+	__get_cpu_var(next_sbit) = (++last >= aisb_max) ? 0 : last;
+}
+
+/* msi_vecs - number of requested interrupts, 0 place function to error state */
+static int zpci_setup_msi(struct pci_dev *pdev, int msi_vecs)
+{
+	struct zpci_dev *zdev = get_zdev(pdev);
+	unsigned int aisb, msi_nr;
+	struct msi_desc *msi;
+	int rc;
+
+	/* store the number of used MSI vectors */
+	zdev->irq_map->msi_vecs = min(msi_vecs, ZPCI_NR_MSI_VECS);
+
+	spin_lock(&bucket->lock);
+	aisb = find_first_zero_bit(bucket->alloc, PAGE_SIZE);
+	/* alloc map exhausted? */
+	if (aisb == PAGE_SIZE) {
+		spin_unlock(&bucket->lock);
+		return -EIO;
+	}
+	set_bit(aisb, bucket->alloc);
+	spin_unlock(&bucket->lock);
+
+	zdev->aisb = aisb;
+	if (aisb + 1 > aisb_max)
+		aisb_max = aisb + 1;
+
+	/* wire up IRQ shortcut pointer */
+	bucket->imap[zdev->aisb] = zdev->irq_map;
+	pr_debug("%s: imap[%u] linked to %p\n", __func__, zdev->aisb, zdev->irq_map);
+
+	/* TODO: irq number 0 wont be found if we return less than requested MSIs.
+	 * ignore it for now and fix in common code.
+	 */
+	msi_nr = aisb << ZPCI_MSI_VEC_BITS;
+
+	list_for_each_entry(msi, &pdev->msi_list, list) {
+		rc = zpci_setup_msi_irq(zdev, msi, msi_nr,
+					  aisb << ZPCI_MSI_VEC_BITS);
+		if (rc)
+			return rc;
+		msi_nr++;
+	}
+
+	rc = zpci_register_airq(zdev, aisb, (u64) &zdev->irq_map->aibv);
+	if (rc) {
+		clear_bit(aisb, bucket->alloc);
+		dev_err(&pdev->dev, "register MSI failed with: %d\n", rc);
+		return rc;
+	}
+	return (zdev->irq_map->msi_vecs == msi_vecs) ?
+		0 : zdev->irq_map->msi_vecs;
+}
+
+static void zpci_teardown_msi(struct pci_dev *pdev)
+{
+	struct zpci_dev *zdev = get_zdev(pdev);
+	struct msi_desc *msi;
+	int aisb, rc;
+
+	rc = zpci_unregister_airq(zdev);
+	if (rc) {
+		dev_err(&pdev->dev, "deregister MSI failed with: %d\n", rc);
+		return;
+	}
+
+	msi = list_first_entry(&pdev->msi_list, struct msi_desc, list);
+	aisb = irq_to_dev_nr(msi->irq);
+
+	list_for_each_entry(msi, &pdev->msi_list, list)
+		zpci_teardown_msi_irq(zdev, msi);
+
+	clear_bit(aisb, bucket->alloc);
+	if (aisb + 1 == aisb_max)
+		aisb_max--;
+}
+
+int arch_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
+{
+	pr_debug("%s: requesting %d MSI-X interrupts...", __func__, nvec);
+	if (type != PCI_CAP_ID_MSIX && type != PCI_CAP_ID_MSI)
+		return -EINVAL;
+	return zpci_setup_msi(pdev, nvec);
+}
+
+void arch_teardown_msi_irqs(struct pci_dev *pdev)
+{
+	pr_info("%s: on pdev: %p\n", __func__, pdev);
+	zpci_teardown_msi(pdev);
+}
+
+static void zpci_map_resources(struct zpci_dev *zdev)
+{
+	struct pci_dev *pdev = zdev->pdev;
+	resource_size_t len;
+	int i;
+
+	for (i = 0; i < PCI_BAR_COUNT; i++) {
+		len = pci_resource_len(pdev, i);
+		if (!len)
+			continue;
+		pdev->resource[i].start = (resource_size_t) pci_iomap(pdev, i, 0);
+		pdev->resource[i].end = pdev->resource[i].start + len - 1;
+		pr_debug("BAR%i: -> start: %Lx  end: %Lx\n",
+			i, pdev->resource[i].start, pdev->resource[i].end);
+	}
+};
+
+static void zpci_unmap_resources(struct pci_dev *pdev)
+{
+	resource_size_t len;
+	int i;
+
+	for (i = 0; i < PCI_BAR_COUNT; i++) {
+		len = pci_resource_len(pdev, i);
+		if (!len)
+			continue;
+		pci_iounmap(pdev, (void *) pdev->resource[i].start);
+	}
+};
+
+struct zpci_dev *zpci_alloc_device(void)
+{
+	struct zpci_dev *zdev;
+
+	/* Alloc memory for our private pci device data */
+	zdev = kzalloc(sizeof(*zdev), GFP_KERNEL);
+	if (!zdev)
+		return ERR_PTR(-ENOMEM);
+
+	/* Alloc aibv & callback space */
+	zdev->irq_map = kmem_cache_zalloc(zdev_irq_cache, GFP_KERNEL);
+	if (!zdev->irq_map)
+		goto error;
+	WARN_ON((u64) zdev->irq_map & 0xff);
+	return zdev;
+
+error:
+	kfree(zdev);
+	return ERR_PTR(-ENOMEM);
+}
+
+void zpci_free_device(struct zpci_dev *zdev)
+{
+	kmem_cache_free(zdev_irq_cache, zdev->irq_map);
+	kfree(zdev);
+}
+
+/* Called on removal of pci_dev, leaves zpci and bus device */
+static void zpci_remove_device(struct pci_dev *pdev)
+{
+	struct zpci_dev *zdev = get_zdev(pdev);
+
+	dev_info(&pdev->dev, "Removing device %u\n", zdev->domain);
+	zdev->state = ZPCI_FN_STATE_CONFIGURED;
+	zpci_dma_exit_device(zdev);
+	zpci_sysfs_remove_device(&pdev->dev);
+	zpci_unmap_resources(pdev);
+	list_del(&zdev->entry);		/* can be called from init */
+	zdev->pdev = NULL;
+}
+
+static void zpci_scan_devices(void)
+{
+	struct zpci_dev *zdev;
+
+	mutex_lock(&zpci_list_lock);
+	list_for_each_entry(zdev, &zpci_list, entry)
+		if (zdev->state == ZPCI_FN_STATE_CONFIGURED)
+			zpci_scan_device(zdev);
+	mutex_unlock(&zpci_list_lock);
+}
+
+/*
+ * Too late for any s390 specific setup, since interrupts must be set up
+ * already which requires DMA setup too and the pci scan will access the
+ * config space, which only works if the function handle is enabled.
+ */
+int pcibios_enable_device(struct pci_dev *pdev, int mask)
+{
+	struct resource *res;
+	u16 cmd;
+	int i;
+
+	pci_read_config_word(pdev, PCI_COMMAND, &cmd);
+
+	for (i = 0; i < PCI_BAR_COUNT; i++) {
+		res = &pdev->resource[i];
+
+		if (res->flags & IORESOURCE_IO)
+			return -EINVAL;
+
+		if (res->flags & IORESOURCE_MEM)
+			cmd |= PCI_COMMAND_MEMORY;
+	}
+	pci_write_config_word(pdev, PCI_COMMAND, cmd);
+	return 0;
+}
+
+void pcibios_disable_device(struct pci_dev *pdev)
+{
+	zpci_remove_device(pdev);
+	pdev->sysdata = NULL;
+}
+
+int pcibios_add_platform_entries(struct pci_dev *pdev)
+{
+	return zpci_sysfs_add_device(&pdev->dev);
+}
+
+int zpci_request_irq(unsigned int irq, irq_handler_t handler, void *data)
+{
+	int msi_nr = irq_to_msi_nr(irq);
+	struct zdev_irq_map *imap;
+	struct msi_desc *msi;
+
+	msi = irq_get_msi_desc(irq);
+	if (!msi)
+		return -EIO;
+
+	imap = get_imap(irq);
+	spin_lock_init(&imap->lock);
+
+	pr_debug("%s: register handler for IRQ:MSI %d:%d\n", __func__, irq >> 6, msi_nr);
+	imap->cb[msi_nr].handler = handler;
+	imap->cb[msi_nr].data = data;
+
+	/*
+	 * The generic MSI code returns with the interrupt disabled on the
+	 * card, using the MSI mask bits. Firmware doesn't appear to unmask
+	 * at that level, so we do it here by hand.
+	 */
+	zpci_msi_set_mask_bits(msi, 1, 0);
+	return 0;
+}
+
+void zpci_free_irq(unsigned int irq)
+{
+	struct zdev_irq_map *imap = get_imap(irq);
+	int msi_nr = irq_to_msi_nr(irq);
+	unsigned long flags;
+
+	pr_debug("%s: for irq: %d\n", __func__, irq);
+
+	spin_lock_irqsave(&imap->lock, flags);
+	imap->cb[msi_nr].handler = NULL;
+	imap->cb[msi_nr].data = NULL;
+	spin_unlock_irqrestore(&imap->lock, flags);
+}
+
+int request_irq(unsigned int irq, irq_handler_t handler,
+		unsigned long irqflags, const char *devname, void *dev_id)
+{
+	pr_debug("%s: irq: %d  handler: %p  flags: %lx  dev: %s\n",
+		__func__, irq, handler, irqflags, devname);
+
+	return zpci_request_irq(irq, handler, dev_id);
+}
+EXPORT_SYMBOL_GPL(request_irq);
+
+void free_irq(unsigned int irq, void *dev_id)
+{
+	zpci_free_irq(irq);
+}
+EXPORT_SYMBOL_GPL(free_irq);
+
+static int __init zpci_irq_init(void)
+{
+	int cpu, rc;
+
+	bucket = kzalloc(sizeof(*bucket), GFP_KERNEL);
+	if (!bucket)
+		return -ENOMEM;
+
+	bucket->aisb = (unsigned long *) get_zeroed_page(GFP_KERNEL);
+	if (!bucket->aisb) {
+		rc = -ENOMEM;
+		goto out_aisb;
+	}
+
+	bucket->alloc = (unsigned long *) get_zeroed_page(GFP_KERNEL);
+	if (!bucket->alloc) {
+		rc = -ENOMEM;
+		goto out_alloc;
+	}
+
+	isc_register(PCI_ISC);
+	zpci_irq_si = s390_register_adapter_interrupt(&zpci_irq_handler, NULL, PCI_ISC);
+	if (IS_ERR(zpci_irq_si)) {
+		rc = PTR_ERR(zpci_irq_si);
+		zpci_irq_si = NULL;
+		goto out_ai;
+	}
+
+	for_each_online_cpu(cpu)
+		per_cpu(next_sbit, cpu) = 0;
+
+	spin_lock_init(&bucket->lock);
+	/* set summary to 1 to be called every time for the ISC */
+	*zpci_irq_si = 1;
+	sic_instr(SIC_IRQ_MODE_SINGLE, NULL, PCI_ISC);
+	return 0;
+
+out_ai:
+	isc_unregister(PCI_ISC);
+	free_page((unsigned long) bucket->alloc);
+out_alloc:
+	free_page((unsigned long) bucket->aisb);
+out_aisb:
+	kfree(bucket);
+	return rc;
+}
+
+static void zpci_irq_exit(void)
+{
+	free_page((unsigned long) bucket->alloc);
+	free_page((unsigned long) bucket->aisb);
+	s390_unregister_adapter_interrupt(zpci_irq_si, PCI_ISC);
+	isc_unregister(PCI_ISC);
+	kfree(bucket);
+}
+
+static struct resource *zpci_alloc_bus_resource(unsigned long start, unsigned long size,
+						unsigned long flags, int domain)
+{
+	struct resource *r;
+	char *name;
+	int rc;
+
+	r = kzalloc(sizeof(*r), GFP_KERNEL);
+	if (!r)
+		return ERR_PTR(-ENOMEM);
+	r->start = start;
+	r->end = r->start + size - 1;
+	r->flags = flags;
+	r->parent = &iomem_resource;
+	name = kmalloc(18, GFP_KERNEL);
+	if (!name) {
+		kfree(r);
+		return ERR_PTR(-ENOMEM);
+	}
+	sprintf(name, "PCI Bus: %04x:%02x", domain, ZPCI_BUS_NR);
+	r->name = name;
+
+	rc = request_resource(&iomem_resource, r);
+	if (rc)
+		pr_debug("request resource %pR failed\n", r);
+	return r;
+}
+
+static int zpci_alloc_iomap(struct zpci_dev *zdev)
+{
+	int entry;
+
+	spin_lock(&zpci_iomap_lock);
+	entry = find_first_zero_bit(zpci_iomap, ZPCI_IOMAP_MAX_ENTRIES);
+	if (entry == ZPCI_IOMAP_MAX_ENTRIES) {
+		spin_unlock(&zpci_iomap_lock);
+		return -ENOSPC;
+	}
+	set_bit(entry, zpci_iomap);
+	spin_unlock(&zpci_iomap_lock);
+	return entry;
+}
+
+static void zpci_free_iomap(struct zpci_dev *zdev, int entry)
+{
+	spin_lock(&zpci_iomap_lock);
+	memset(&zpci_iomap_start[entry], 0, sizeof(struct zpci_iomap_entry));
+	clear_bit(entry, zpci_iomap);
+	spin_unlock(&zpci_iomap_lock);
+}
+
+static int zpci_create_device_bus(struct zpci_dev *zdev)
+{
+	struct resource *res;
+	LIST_HEAD(resources);
+	int i;
+
+	/* allocate mapping entry for each used bar */
+	for (i = 0; i < PCI_BAR_COUNT; i++) {
+		unsigned long addr, size, flags;
+		int entry;
+
+		if (!zdev->bars[i].size)
+			continue;
+		entry = zpci_alloc_iomap(zdev);
+		if (entry < 0)
+			return entry;
+		zdev->bars[i].map_idx = entry;
+
+		/* only MMIO is supported */
+		flags = IORESOURCE_MEM;
+		if (zdev->bars[i].val & 8)
+			flags |= IORESOURCE_PREFETCH;
+		if (zdev->bars[i].val & 4)
+			flags |= IORESOURCE_MEM_64;
+
+		addr = ZPCI_IOMAP_ADDR_BASE + ((u64) entry << 48);
+
+		size = 1UL << zdev->bars[i].size;
+
+		res = zpci_alloc_bus_resource(addr, size, flags, zdev->domain);
+		if (IS_ERR(res)) {
+			zpci_free_iomap(zdev, entry);
+			return PTR_ERR(res);
+		}
+		pci_add_resource(&resources, res);
+	}
+
+	zdev->bus = pci_create_root_bus(NULL, ZPCI_BUS_NR, &pci_root_ops,
+					zdev, &resources);
+	if (!zdev->bus)
+		return -EIO;
+
+	zdev->bus->max_bus_speed = zdev->max_bus_speed;
+	return 0;
+}
+
+static int zpci_alloc_domain(struct zpci_dev *zdev)
+{
+	spin_lock(&zpci_domain_lock);
+	zdev->domain = find_first_zero_bit(zpci_domain, ZPCI_NR_DEVICES);
+	if (zdev->domain == ZPCI_NR_DEVICES) {
+		spin_unlock(&zpci_domain_lock);
+		return -ENOSPC;
+	}
+	set_bit(zdev->domain, zpci_domain);
+	spin_unlock(&zpci_domain_lock);
+	return 0;
+}
+
+static void zpci_free_domain(struct zpci_dev *zdev)
+{
+	spin_lock(&zpci_domain_lock);
+	clear_bit(zdev->domain, zpci_domain);
+	spin_unlock(&zpci_domain_lock);
+}
+
+int zpci_enable_device(struct zpci_dev *zdev)
+{
+	int rc;
+
+	rc = clp_enable_fh(zdev, ZPCI_NR_DMA_SPACES);
+	if (rc)
+		goto out;
+	pr_info("Enabled fh: 0x%x fid: 0x%x\n", zdev->fh, zdev->fid);
+
+	rc = zpci_dma_init_device(zdev);
+	if (rc)
+		goto out_dma;
+	return 0;
+
+out_dma:
+	clp_disable_fh(zdev);
+out:
+	return rc;
+}
+EXPORT_SYMBOL_GPL(zpci_enable_device);
+
+int zpci_create_device(struct zpci_dev *zdev)
+{
+	int rc;
+
+	rc = zpci_alloc_domain(zdev);
+	if (rc)
+		goto out;
+
+	rc = zpci_create_device_bus(zdev);
+	if (rc)
+		goto out_bus;
+
+	mutex_lock(&zpci_list_lock);
+	list_add_tail(&zdev->entry, &zpci_list);
+	if (hotplug_ops.create_slot)
+		hotplug_ops.create_slot(zdev);
+	mutex_unlock(&zpci_list_lock);
+
+	if (zdev->state == ZPCI_FN_STATE_STANDBY)
+		return 0;
+
+	rc = zpci_enable_device(zdev);
+	if (rc)
+		goto out_start;
+	return 0;
+
+out_start:
+	mutex_lock(&zpci_list_lock);
+	list_del(&zdev->entry);
+	if (hotplug_ops.remove_slot)
+		hotplug_ops.remove_slot(zdev);
+	mutex_unlock(&zpci_list_lock);
+out_bus:
+	zpci_free_domain(zdev);
+out:
+	return rc;
+}
+
+void zpci_stop_device(struct zpci_dev *zdev)
+{
+	zpci_dma_exit_device(zdev);
+	/*
+	 * Note: SCLP disables fh via set-pci-fn so don't
+	 * do that here.
+	 */
+}
+EXPORT_SYMBOL_GPL(zpci_stop_device);
+
+int zpci_scan_device(struct zpci_dev *zdev)
+{
+	zdev->pdev = pci_scan_single_device(zdev->bus, ZPCI_DEVFN);
+	if (!zdev->pdev) {
+		pr_err("pci_scan_single_device failed for fid: 0x%x\n",
+			zdev->fid);
+		goto out;
+	}
+
+	zpci_map_resources(zdev);
+	pci_bus_add_devices(zdev->bus);
+
+	/* now that pdev was added to the bus mark it as used */
+	zdev->state = ZPCI_FN_STATE_ONLINE;
+	return 0;
+
+out:
+	zpci_dma_exit_device(zdev);
+	clp_disable_fh(zdev);
+	return -EIO;
+}
+EXPORT_SYMBOL_GPL(zpci_scan_device);
+
+static inline int barsize(u8 size)
+{
+	return (size) ? (1 << size) >> 10 : 0;
+}
+
+static int zpci_mem_init(void)
+{
+	zdev_irq_cache = kmem_cache_create("PCI_IRQ_cache", sizeof(struct zdev_irq_map),
+				L1_CACHE_BYTES, SLAB_HWCACHE_ALIGN, NULL);
+	if (!zdev_irq_cache)
+		goto error_zdev;
+
+	/* TODO: use realloc */
+	zpci_iomap_start = kzalloc(ZPCI_IOMAP_MAX_ENTRIES * sizeof(*zpci_iomap_start),
+				   GFP_KERNEL);
+	if (!zpci_iomap_start)
+		goto error_iomap;
+	return 0;
+
+error_iomap:
+	kmem_cache_destroy(zdev_irq_cache);
+error_zdev:
+	return -ENOMEM;
+}
+
+static void zpci_mem_exit(void)
+{
+	kfree(zpci_iomap_start);
+	kmem_cache_destroy(zdev_irq_cache);
+}
+
+unsigned int pci_probe = 1;
+EXPORT_SYMBOL_GPL(pci_probe);
+
+char * __init pcibios_setup(char *str)
+{
+	if (!strcmp(str, "off")) {
+		pci_probe = 0;
+		return NULL;
+	}
+	return str;
+}
+
+static int __init pci_base_init(void)
+{
+	int rc;
+
+	if (!pci_probe)
+		return 0;
+
+	if (!test_facility(2) || !test_facility(69)
+	    || !test_facility(71) || !test_facility(72))
+		return 0;
+
+	pr_info("Probing PCI hardware: PCI:%d  SID:%d  AEN:%d\n",
+		test_facility(69), test_facility(70),
+		test_facility(71));
+
+	rc = zpci_mem_init();
+	if (rc)
+		goto out_mem;
+
+	rc = zpci_msihash_init();
+	if (rc)
+		goto out_hash;
+
+	rc = zpci_irq_init();
+	if (rc)
+		goto out_irq;
+
+	rc = zpci_dma_init();
+	if (rc)
+		goto out_dma;
+
+	rc = clp_find_pci_devices();
+	if (rc)
+		goto out_find;
+
+	zpci_scan_devices();
+	return 0;
+
+out_find:
+	zpci_dma_exit();
+out_dma:
+	zpci_irq_exit();
+out_irq:
+	zpci_msihash_exit();
+out_hash:
+	zpci_mem_exit();
+out_mem:
+	return rc;
+}
+subsys_initcall(pci_base_init);
diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c
new file mode 100644
index 000000000000..7f4ce8d874a4
--- /dev/null
+++ b/arch/s390/pci/pci_clp.c
@@ -0,0 +1,324 @@
+/*
+ * Copyright IBM Corp. 2012
+ *
+ * Author(s):
+ *   Jan Glauber <jang@linux.vnet.ibm.com>
+ */
+
+#define COMPONENT "zPCI"
+#define pr_fmt(fmt) COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/err.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <asm/pci_clp.h>
+
+/*
+ * Call Logical Processor
+ * Retry logic is handled by the caller.
+ */
+static inline u8 clp_instr(void *req)
+{
+	u64 ilpm;
+	u8 cc;
+
+	asm volatile (
+		"	.insn	rrf,0xb9a00000,%[ilpm],%[req],0x0,0x2\n"
+		"	ipm	%[cc]\n"
+		"	srl	%[cc],28\n"
+		: [cc] "=d" (cc), [ilpm] "=d" (ilpm)
+		: [req] "a" (req)
+		: "cc", "memory");
+	return cc;
+}
+
+static void *clp_alloc_block(void)
+{
+	struct page *page = alloc_pages(GFP_KERNEL, get_order(CLP_BLK_SIZE));
+	return (page) ? page_address(page) : NULL;
+}
+
+static void clp_free_block(void *ptr)
+{
+	free_pages((unsigned long) ptr, get_order(CLP_BLK_SIZE));
+}
+
+static void clp_store_query_pci_fngrp(struct zpci_dev *zdev,
+				      struct clp_rsp_query_pci_grp *response)
+{
+	zdev->tlb_refresh = response->refresh;
+	zdev->dma_mask = response->dasm;
+	zdev->msi_addr = response->msia;
+
+	pr_debug("Supported number of MSI vectors: %u\n", response->noi);
+	switch (response->version) {
+	case 1:
+		zdev->max_bus_speed = PCIE_SPEED_5_0GT;
+		break;
+	default:
+		zdev->max_bus_speed = PCI_SPEED_UNKNOWN;
+		break;
+	}
+}
+
+static int clp_query_pci_fngrp(struct zpci_dev *zdev, u8 pfgid)
+{
+	struct clp_req_rsp_query_pci_grp *rrb;
+	int rc;
+
+	rrb = clp_alloc_block();
+	if (!rrb)
+		return -ENOMEM;
+
+	memset(rrb, 0, sizeof(*rrb));
+	rrb->request.hdr.len = sizeof(rrb->request);
+	rrb->request.hdr.cmd = CLP_QUERY_PCI_FNGRP;
+	rrb->response.hdr.len = sizeof(rrb->response);
+	rrb->request.pfgid = pfgid;
+
+	rc = clp_instr(rrb);
+	if (!rc && rrb->response.hdr.rsp == CLP_RC_OK)
+		clp_store_query_pci_fngrp(zdev, &rrb->response);
+	else {
+		pr_err("Query PCI FNGRP failed with response: %x  cc: %d\n",
+			rrb->response.hdr.rsp, rc);
+		rc = -EIO;
+	}
+	clp_free_block(rrb);
+	return rc;
+}
+
+static int clp_store_query_pci_fn(struct zpci_dev *zdev,
+				  struct clp_rsp_query_pci *response)
+{
+	int i;
+
+	for (i = 0; i < PCI_BAR_COUNT; i++) {
+		zdev->bars[i].val = le32_to_cpu(response->bar[i]);
+		zdev->bars[i].size = response->bar_size[i];
+	}
+	zdev->start_dma = response->sdma;
+	zdev->end_dma = response->edma;
+	zdev->pchid = response->pchid;
+	zdev->pfgid = response->pfgid;
+	return 0;
+}
+
+static int clp_query_pci_fn(struct zpci_dev *zdev, u32 fh)
+{
+	struct clp_req_rsp_query_pci *rrb;
+	int rc;
+
+	rrb = clp_alloc_block();
+	if (!rrb)
+		return -ENOMEM;
+
+	memset(rrb, 0, sizeof(*rrb));
+	rrb->request.hdr.len = sizeof(rrb->request);
+	rrb->request.hdr.cmd = CLP_QUERY_PCI_FN;
+	rrb->response.hdr.len = sizeof(rrb->response);
+	rrb->request.fh = fh;
+
+	rc = clp_instr(rrb);
+	if (!rc && rrb->response.hdr.rsp == CLP_RC_OK) {
+		rc = clp_store_query_pci_fn(zdev, &rrb->response);
+		if (rc)
+			goto out;
+		if (rrb->response.pfgid)
+			rc = clp_query_pci_fngrp(zdev, rrb->response.pfgid);
+	} else {
+		pr_err("Query PCI failed with response: %x  cc: %d\n",
+			 rrb->response.hdr.rsp, rc);
+		rc = -EIO;
+	}
+out:
+	clp_free_block(rrb);
+	return rc;
+}
+
+int clp_add_pci_device(u32 fid, u32 fh, int configured)
+{
+	struct zpci_dev *zdev;
+	int rc;
+
+	zdev = zpci_alloc_device();
+	if (IS_ERR(zdev))
+		return PTR_ERR(zdev);
+
+	zdev->fh = fh;
+	zdev->fid = fid;
+
+	/* Query function properties and update zdev */
+	rc = clp_query_pci_fn(zdev, fh);
+	if (rc)
+		goto error;
+
+	if (configured)
+		zdev->state = ZPCI_FN_STATE_CONFIGURED;
+	else
+		zdev->state = ZPCI_FN_STATE_STANDBY;
+
+	rc = zpci_create_device(zdev);
+	if (rc)
+		goto error;
+	return 0;
+
+error:
+	zpci_free_device(zdev);
+	return rc;
+}
+
+/*
+ * Enable/Disable a given PCI function defined by its function handle.
+ */
+static int clp_set_pci_fn(u32 *fh, u8 nr_dma_as, u8 command)
+{
+	struct clp_req_rsp_set_pci *rrb;
+	int rc, retries = 1000;
+
+	rrb = clp_alloc_block();
+	if (!rrb)
+		return -ENOMEM;
+
+	do {
+		memset(rrb, 0, sizeof(*rrb));
+		rrb->request.hdr.len = sizeof(rrb->request);
+		rrb->request.hdr.cmd = CLP_SET_PCI_FN;
+		rrb->response.hdr.len = sizeof(rrb->response);
+		rrb->request.fh = *fh;
+		rrb->request.oc = command;
+		rrb->request.ndas = nr_dma_as;
+
+		rc = clp_instr(rrb);
+		if (rrb->response.hdr.rsp == CLP_RC_SETPCIFN_BUSY) {
+			retries--;
+			if (retries < 0)
+				break;
+			msleep(1);
+		}
+	} while (rrb->response.hdr.rsp == CLP_RC_SETPCIFN_BUSY);
+
+	if (!rc && rrb->response.hdr.rsp == CLP_RC_OK)
+		*fh = rrb->response.fh;
+	else {
+		pr_err("Set PCI FN failed with response: %x  cc: %d\n",
+			rrb->response.hdr.rsp, rc);
+		rc = -EIO;
+	}
+	clp_free_block(rrb);
+	return rc;
+}
+
+int clp_enable_fh(struct zpci_dev *zdev, u8 nr_dma_as)
+{
+	u32 fh = zdev->fh;
+	int rc;
+
+	rc = clp_set_pci_fn(&fh, nr_dma_as, CLP_SET_ENABLE_PCI_FN);
+	if (!rc)
+		/* Success -> store enabled handle in zdev */
+		zdev->fh = fh;
+	return rc;
+}
+
+int clp_disable_fh(struct zpci_dev *zdev)
+{
+	u32 fh = zdev->fh;
+	int rc;
+
+	if (!zdev_enabled(zdev))
+		return 0;
+
+	dev_info(&zdev->pdev->dev, "disabling fn handle: 0x%x\n", fh);
+	rc = clp_set_pci_fn(&fh, 0, CLP_SET_DISABLE_PCI_FN);
+	if (!rc)
+		/* Success -> store disabled handle in zdev */
+		zdev->fh = fh;
+	else
+		dev_err(&zdev->pdev->dev,
+			"Failed to disable fn handle: 0x%x\n", fh);
+	return rc;
+}
+
+static void clp_check_pcifn_entry(struct clp_fh_list_entry *entry)
+{
+	int present, rc;
+
+	if (!entry->vendor_id)
+		return;
+
+	/* TODO: be a little bit more scalable */
+	present = zpci_fid_present(entry->fid);
+
+	if (present)
+		pr_debug("%s: device %x already present\n", __func__, entry->fid);
+
+	/* skip already used functions */
+	if (present && entry->config_state)
+		return;
+
+	/* aev 306: function moved to stand-by state */
+	if (present && !entry->config_state) {
+		/*
+		 * The handle is already disabled, that means no iota/irq freeing via
+		 * the firmware interfaces anymore. Need to free resources manually
+		 * (DMA memory, debug, sysfs)...
+		 */
+		zpci_stop_device(get_zdev_by_fid(entry->fid));
+		return;
+	}
+
+	rc = clp_add_pci_device(entry->fid, entry->fh, entry->config_state);
+	if (rc)
+		pr_err("Failed to add fid: 0x%x\n", entry->fid);
+}
+
+int clp_find_pci_devices(void)
+{
+	struct clp_req_rsp_list_pci *rrb;
+	u64 resume_token = 0;
+	int entries, i, rc;
+
+	rrb = clp_alloc_block();
+	if (!rrb)
+		return -ENOMEM;
+
+	do {
+		memset(rrb, 0, sizeof(*rrb));
+		rrb->request.hdr.len = sizeof(rrb->request);
+		rrb->request.hdr.cmd = CLP_LIST_PCI;
+		/* store as many entries as possible */
+		rrb->response.hdr.len = CLP_BLK_SIZE - LIST_PCI_HDR_LEN;
+		rrb->request.resume_token = resume_token;
+
+		/* Get PCI function handle list */
+		rc = clp_instr(rrb);
+		if (rc || rrb->response.hdr.rsp != CLP_RC_OK) {
+			pr_err("List PCI failed with response: 0x%x  cc: %d\n",
+				rrb->response.hdr.rsp, rc);
+			rc = -EIO;
+			goto out;
+		}
+
+		WARN_ON_ONCE(rrb->response.entry_size !=
+			sizeof(struct clp_fh_list_entry));
+
+		entries = (rrb->response.hdr.len - LIST_PCI_HDR_LEN) /
+			rrb->response.entry_size;
+		pr_info("Detected number of PCI functions: %u\n", entries);
+
+		/* Store the returned resume token as input for the next call */
+		resume_token = rrb->response.resume_token;
+
+		for (i = 0; i < entries; i++)
+			clp_check_pcifn_entry(&rrb->response.fh_list[i]);
+	} while (resume_token);
+
+	pr_debug("Maximum number of supported PCI functions: %u\n",
+		rrb->response.max_fn);
+out:
+	clp_free_block(rrb);
+	return rc;
+}
diff --git a/arch/s390/pci/pci_dma.c b/arch/s390/pci/pci_dma.c
new file mode 100644
index 000000000000..c64b4b294b0a
--- /dev/null
+++ b/arch/s390/pci/pci_dma.c
@@ -0,0 +1,506 @@
+/*
+ * Copyright IBM Corp. 2012
+ *
+ * Author(s):
+ *   Jan Glauber <jang@linux.vnet.ibm.com>
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+#include <linux/iommu-helper.h>
+#include <linux/dma-mapping.h>
+#include <linux/pci.h>
+#include <asm/pci_dma.h>
+
+static enum zpci_ioat_dtype zpci_ioat_dt = ZPCI_IOTA_RTTO;
+
+static struct kmem_cache *dma_region_table_cache;
+static struct kmem_cache *dma_page_table_cache;
+
+static unsigned long *dma_alloc_cpu_table(void)
+{
+	unsigned long *table, *entry;
+
+	table = kmem_cache_alloc(dma_region_table_cache, GFP_ATOMIC);
+	if (!table)
+		return NULL;
+
+	for (entry = table; entry < table + ZPCI_TABLE_ENTRIES; entry++)
+		*entry = ZPCI_TABLE_INVALID | ZPCI_TABLE_PROTECTED;
+	return table;
+}
+
+static void dma_free_cpu_table(void *table)
+{
+	kmem_cache_free(dma_region_table_cache, table);
+}
+
+static unsigned long *dma_alloc_page_table(void)
+{
+	unsigned long *table, *entry;
+
+	table = kmem_cache_alloc(dma_page_table_cache, GFP_ATOMIC);
+	if (!table)
+		return NULL;
+
+	for (entry = table; entry < table + ZPCI_PT_ENTRIES; entry++)
+		*entry = ZPCI_PTE_INVALID | ZPCI_TABLE_PROTECTED;
+	return table;
+}
+
+static void dma_free_page_table(void *table)
+{
+	kmem_cache_free(dma_page_table_cache, table);
+}
+
+static unsigned long *dma_get_seg_table_origin(unsigned long *entry)
+{
+	unsigned long *sto;
+
+	if (reg_entry_isvalid(*entry))
+		sto = get_rt_sto(*entry);
+	else {
+		sto = dma_alloc_cpu_table();
+		if (!sto)
+			return NULL;
+
+		set_rt_sto(entry, sto);
+		validate_rt_entry(entry);
+		entry_clr_protected(entry);
+	}
+	return sto;
+}
+
+static unsigned long *dma_get_page_table_origin(unsigned long *entry)
+{
+	unsigned long *pto;
+
+	if (reg_entry_isvalid(*entry))
+		pto = get_st_pto(*entry);
+	else {
+		pto = dma_alloc_page_table();
+		if (!pto)
+			return NULL;
+		set_st_pto(entry, pto);
+		validate_st_entry(entry);
+		entry_clr_protected(entry);
+	}
+	return pto;
+}
+
+static unsigned long *dma_walk_cpu_trans(unsigned long *rto, dma_addr_t dma_addr)
+{
+	unsigned long *sto, *pto;
+	unsigned int rtx, sx, px;
+
+	rtx = calc_rtx(dma_addr);
+	sto = dma_get_seg_table_origin(&rto[rtx]);
+	if (!sto)
+		return NULL;
+
+	sx = calc_sx(dma_addr);
+	pto = dma_get_page_table_origin(&sto[sx]);
+	if (!pto)
+		return NULL;
+
+	px = calc_px(dma_addr);
+	return &pto[px];
+}
+
+static void dma_update_cpu_trans(struct zpci_dev *zdev, void *page_addr,
+				 dma_addr_t dma_addr, int flags)
+{
+	unsigned long *entry;
+
+	entry = dma_walk_cpu_trans(zdev->dma_table, dma_addr);
+	if (!entry) {
+		WARN_ON_ONCE(1);
+		return;
+	}
+
+	if (flags & ZPCI_PTE_INVALID) {
+		invalidate_pt_entry(entry);
+		return;
+	} else {
+		set_pt_pfaa(entry, page_addr);
+		validate_pt_entry(entry);
+	}
+
+	if (flags & ZPCI_TABLE_PROTECTED)
+		entry_set_protected(entry);
+	else
+		entry_clr_protected(entry);
+}
+
+static int dma_update_trans(struct zpci_dev *zdev, unsigned long pa,
+			    dma_addr_t dma_addr, size_t size, int flags)
+{
+	unsigned int nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	u8 *page_addr = (u8 *) (pa & PAGE_MASK);
+	dma_addr_t start_dma_addr = dma_addr;
+	unsigned long irq_flags;
+	int i, rc = 0;
+
+	if (!nr_pages)
+		return -EINVAL;
+
+	spin_lock_irqsave(&zdev->dma_table_lock, irq_flags);
+	if (!zdev->dma_table) {
+		dev_err(&zdev->pdev->dev, "Missing DMA table\n");
+		goto no_refresh;
+	}
+
+	for (i = 0; i < nr_pages; i++) {
+		dma_update_cpu_trans(zdev, page_addr, dma_addr, flags);
+		page_addr += PAGE_SIZE;
+		dma_addr += PAGE_SIZE;
+	}
+
+	/*
+	 * rpcit is not required to establish new translations when previously
+	 * invalid translation-table entries are validated, however it is
+	 * required when altering previously valid entries.
+	 */
+	if (!zdev->tlb_refresh &&
+	    ((flags & ZPCI_PTE_VALID_MASK) == ZPCI_PTE_VALID))
+		/*
+		 * TODO: also need to check that the old entry is indeed INVALID
+		 * and not only for one page but for the whole range...
+		 * -> now we WARN_ON in that case but with lazy unmap that
+		 * needs to be redone!
+		 */
+		goto no_refresh;
+	rc = rpcit_instr((u64) zdev->fh << 32, start_dma_addr,
+			  nr_pages * PAGE_SIZE);
+
+no_refresh:
+	spin_unlock_irqrestore(&zdev->dma_table_lock, irq_flags);
+	return rc;
+}
+
+static void dma_free_seg_table(unsigned long entry)
+{
+	unsigned long *sto = get_rt_sto(entry);
+	int sx;
+
+	for (sx = 0; sx < ZPCI_TABLE_ENTRIES; sx++)
+		if (reg_entry_isvalid(sto[sx]))
+			dma_free_page_table(get_st_pto(sto[sx]));
+
+	dma_free_cpu_table(sto);
+}
+
+static void dma_cleanup_tables(struct zpci_dev *zdev)
+{
+	unsigned long *table;
+	int rtx;
+
+	if (!zdev || !zdev->dma_table)
+		return;
+
+	table = zdev->dma_table;
+	for (rtx = 0; rtx < ZPCI_TABLE_ENTRIES; rtx++)
+		if (reg_entry_isvalid(table[rtx]))
+			dma_free_seg_table(table[rtx]);
+
+	dma_free_cpu_table(table);
+	zdev->dma_table = NULL;
+}
+
+static unsigned long __dma_alloc_iommu(struct zpci_dev *zdev, unsigned long start,
+				   int size)
+{
+	unsigned long boundary_size = 0x1000000;
+
+	return iommu_area_alloc(zdev->iommu_bitmap, zdev->iommu_pages,
+				start, size, 0, boundary_size, 0);
+}
+
+static unsigned long dma_alloc_iommu(struct zpci_dev *zdev, int size)
+{
+	unsigned long offset, flags;
+
+	spin_lock_irqsave(&zdev->iommu_bitmap_lock, flags);
+	offset = __dma_alloc_iommu(zdev, zdev->next_bit, size);
+	if (offset == -1)
+		offset = __dma_alloc_iommu(zdev, 0, size);
+
+	if (offset != -1) {
+		zdev->next_bit = offset + size;
+		if (zdev->next_bit >= zdev->iommu_pages)
+			zdev->next_bit = 0;
+	}
+	spin_unlock_irqrestore(&zdev->iommu_bitmap_lock, flags);
+	return offset;
+}
+
+static void dma_free_iommu(struct zpci_dev *zdev, unsigned long offset, int size)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&zdev->iommu_bitmap_lock, flags);
+	if (!zdev->iommu_bitmap)
+		goto out;
+	bitmap_clear(zdev->iommu_bitmap, offset, size);
+	if (offset >= zdev->next_bit)
+		zdev->next_bit = offset + size;
+out:
+	spin_unlock_irqrestore(&zdev->iommu_bitmap_lock, flags);
+}
+
+int dma_set_mask(struct device *dev, u64 mask)
+{
+	if (!dev->dma_mask || !dma_supported(dev, mask))
+		return -EIO;
+
+	*dev->dma_mask = mask;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(dma_set_mask);
+
+static dma_addr_t s390_dma_map_pages(struct device *dev, struct page *page,
+				     unsigned long offset, size_t size,
+				     enum dma_data_direction direction,
+				     struct dma_attrs *attrs)
+{
+	struct zpci_dev *zdev = get_zdev(container_of(dev, struct pci_dev, dev));
+	unsigned long nr_pages, iommu_page_index;
+	unsigned long pa = page_to_phys(page) + offset;
+	int flags = ZPCI_PTE_VALID;
+	dma_addr_t dma_addr;
+
+	WARN_ON_ONCE(offset > PAGE_SIZE);
+
+	/* This rounds up number of pages based on size and offset */
+	nr_pages = iommu_num_pages(pa, size, PAGE_SIZE);
+	iommu_page_index = dma_alloc_iommu(zdev, nr_pages);
+	if (iommu_page_index == -1)
+		goto out_err;
+
+	/* Use rounded up size */
+	size = nr_pages * PAGE_SIZE;
+
+	dma_addr = zdev->start_dma + iommu_page_index * PAGE_SIZE;
+	if (dma_addr + size > zdev->end_dma) {
+		dev_err(dev, "(dma_addr: 0x%16.16LX + size: 0x%16.16lx) > end_dma: 0x%16.16Lx\n",
+			 dma_addr, size, zdev->end_dma);
+		goto out_free;
+	}
+
+	if (direction == DMA_NONE || direction == DMA_TO_DEVICE)
+		flags |= ZPCI_TABLE_PROTECTED;
+
+	if (!dma_update_trans(zdev, pa, dma_addr, size, flags))
+		return dma_addr + offset;
+
+out_free:
+	dma_free_iommu(zdev, iommu_page_index, nr_pages);
+out_err:
+	dev_err(dev, "Failed to map addr: %lx\n", pa);
+	return DMA_ERROR_CODE;
+}
+
+static void s390_dma_unmap_pages(struct device *dev, dma_addr_t dma_addr,
+				 size_t size, enum dma_data_direction direction,
+				 struct dma_attrs *attrs)
+{
+	struct zpci_dev *zdev = get_zdev(container_of(dev, struct pci_dev, dev));
+	unsigned long iommu_page_index;
+	int npages;
+
+	npages = iommu_num_pages(dma_addr, size, PAGE_SIZE);
+	dma_addr = dma_addr & PAGE_MASK;
+	if (dma_update_trans(zdev, 0, dma_addr, npages * PAGE_SIZE,
+			     ZPCI_TABLE_PROTECTED | ZPCI_PTE_INVALID))
+		dev_err(dev, "Failed to unmap addr: %Lx\n", dma_addr);
+
+	iommu_page_index = (dma_addr - zdev->start_dma) >> PAGE_SHIFT;
+	dma_free_iommu(zdev, iommu_page_index, npages);
+}
+
+static void *s390_dma_alloc(struct device *dev, size_t size,
+			    dma_addr_t *dma_handle, gfp_t flag,
+			    struct dma_attrs *attrs)
+{
+	struct page *page;
+	unsigned long pa;
+	dma_addr_t map;
+
+	size = PAGE_ALIGN(size);
+	page = alloc_pages(flag, get_order(size));
+	if (!page)
+		return NULL;
+	pa = page_to_phys(page);
+	memset((void *) pa, 0, size);
+
+	map = s390_dma_map_pages(dev, page, pa % PAGE_SIZE,
+				 size, DMA_BIDIRECTIONAL, NULL);
+	if (dma_mapping_error(dev, map)) {
+		free_pages(pa, get_order(size));
+		return NULL;
+	}
+
+	if (dma_handle)
+		*dma_handle = map;
+	return (void *) pa;
+}
+
+static void s390_dma_free(struct device *dev, size_t size,
+			  void *pa, dma_addr_t dma_handle,
+			  struct dma_attrs *attrs)
+{
+	s390_dma_unmap_pages(dev, dma_handle, PAGE_ALIGN(size),
+			     DMA_BIDIRECTIONAL, NULL);
+	free_pages((unsigned long) pa, get_order(size));
+}
+
+static int s390_dma_map_sg(struct device *dev, struct scatterlist *sg,
+			   int nr_elements, enum dma_data_direction dir,
+			   struct dma_attrs *attrs)
+{
+	int mapped_elements = 0;
+	struct scatterlist *s;
+	int i;
+
+	for_each_sg(sg, s, nr_elements, i) {
+		struct page *page = sg_page(s);
+		s->dma_address = s390_dma_map_pages(dev, page, s->offset,
+						    s->length, dir, NULL);
+		if (!dma_mapping_error(dev, s->dma_address)) {
+			s->dma_length = s->length;
+			mapped_elements++;
+		} else
+			goto unmap;
+	}
+out:
+	return mapped_elements;
+
+unmap:
+	for_each_sg(sg, s, mapped_elements, i) {
+		if (s->dma_address)
+			s390_dma_unmap_pages(dev, s->dma_address, s->dma_length,
+					     dir, NULL);
+		s->dma_address = 0;
+		s->dma_length = 0;
+	}
+	mapped_elements = 0;
+	goto out;
+}
+
+static void s390_dma_unmap_sg(struct device *dev, struct scatterlist *sg,
+			      int nr_elements, enum dma_data_direction dir,
+			      struct dma_attrs *attrs)
+{
+	struct scatterlist *s;
+	int i;
+
+	for_each_sg(sg, s, nr_elements, i) {
+		s390_dma_unmap_pages(dev, s->dma_address, s->dma_length, dir, NULL);
+		s->dma_address = 0;
+		s->dma_length = 0;
+	}
+}
+
+int zpci_dma_init_device(struct zpci_dev *zdev)
+{
+	unsigned int bitmap_order;
+	int rc;
+
+	spin_lock_init(&zdev->iommu_bitmap_lock);
+	spin_lock_init(&zdev->dma_table_lock);
+
+	zdev->dma_table = dma_alloc_cpu_table();
+	if (!zdev->dma_table) {
+		rc = -ENOMEM;
+		goto out_clean;
+	}
+
+	zdev->iommu_size = (unsigned long) high_memory - PAGE_OFFSET;
+	zdev->iommu_pages = zdev->iommu_size >> PAGE_SHIFT;
+	bitmap_order = get_order(zdev->iommu_pages / 8);
+	pr_info("iommu_size: 0x%lx  iommu_pages: 0x%lx  bitmap_order: %i\n",
+		 zdev->iommu_size, zdev->iommu_pages, bitmap_order);
+
+	zdev->iommu_bitmap = (void *) __get_free_pages(GFP_KERNEL | __GFP_ZERO,
+						       bitmap_order);
+	if (!zdev->iommu_bitmap) {
+		rc = -ENOMEM;
+		goto out_reg;
+	}
+
+	rc = zpci_register_ioat(zdev,
+				0,
+				zdev->start_dma + PAGE_OFFSET,
+				zdev->start_dma + zdev->iommu_size - 1,
+				(u64) zdev->dma_table);
+	if (rc)
+		goto out_reg;
+	return 0;
+
+out_reg:
+	dma_free_cpu_table(zdev->dma_table);
+out_clean:
+	return rc;
+}
+
+void zpci_dma_exit_device(struct zpci_dev *zdev)
+{
+	zpci_unregister_ioat(zdev, 0);
+	dma_cleanup_tables(zdev);
+	free_pages((unsigned long) zdev->iommu_bitmap,
+		   get_order(zdev->iommu_pages / 8));
+	zdev->iommu_bitmap = NULL;
+	zdev->next_bit = 0;
+}
+
+static int __init dma_alloc_cpu_table_caches(void)
+{
+	dma_region_table_cache = kmem_cache_create("PCI_DMA_region_tables",
+					ZPCI_TABLE_SIZE, ZPCI_TABLE_ALIGN,
+					0, NULL);
+	if (!dma_region_table_cache)
+		return -ENOMEM;
+
+	dma_page_table_cache = kmem_cache_create("PCI_DMA_page_tables",
+					ZPCI_PT_SIZE, ZPCI_PT_ALIGN,
+					0, NULL);
+	if (!dma_page_table_cache) {
+		kmem_cache_destroy(dma_region_table_cache);
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+int __init zpci_dma_init(void)
+{
+	return dma_alloc_cpu_table_caches();
+}
+
+void zpci_dma_exit(void)
+{
+	kmem_cache_destroy(dma_page_table_cache);
+	kmem_cache_destroy(dma_region_table_cache);
+}
+
+#define PREALLOC_DMA_DEBUG_ENTRIES	(1 << 16)
+
+static int __init dma_debug_do_init(void)
+{
+	dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES);
+	return 0;
+}
+fs_initcall(dma_debug_do_init);
+
+struct dma_map_ops s390_dma_ops = {
+	.alloc		= s390_dma_alloc,
+	.free		= s390_dma_free,
+	.map_sg		= s390_dma_map_sg,
+	.unmap_sg	= s390_dma_unmap_sg,
+	.map_page	= s390_dma_map_pages,
+	.unmap_page	= s390_dma_unmap_pages,
+	/* if we support direct DMA this must be conditional */
+	.is_phys	= 0,
+	/* dma_supported is unconditionally true without a callback */
+};
+EXPORT_SYMBOL_GPL(s390_dma_ops);
diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c
new file mode 100644
index 000000000000..dbed8cd3370c
--- /dev/null
+++ b/arch/s390/pci/pci_event.c
@@ -0,0 +1,93 @@
+/*
+ *  Copyright IBM Corp. 2012
+ *
+ *  Author(s):
+ *    Jan Glauber <jang@linux.vnet.ibm.com>
+ */
+
+#define COMPONENT "zPCI"
+#define pr_fmt(fmt) COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+
+/* Content Code Description for PCI Function Error */
+struct zpci_ccdf_err {
+	u32 reserved1;
+	u32 fh;				/* function handle */
+	u32 fid;			/* function id */
+	u32 ett		:  4;		/* expected table type */
+	u32 mvn		: 12;		/* MSI vector number */
+	u32 dmaas	:  8;		/* DMA address space */
+	u32		:  6;
+	u32 q		:  1;		/* event qualifier */
+	u32 rw		:  1;		/* read/write */
+	u64 faddr;			/* failing address */
+	u32 reserved3;
+	u16 reserved4;
+	u16 pec;			/* PCI event code */
+} __packed;
+
+/* Content Code Description for PCI Function Availability */
+struct zpci_ccdf_avail {
+	u32 reserved1;
+	u32 fh;				/* function handle */
+	u32 fid;			/* function id */
+	u32 reserved2;
+	u32 reserved3;
+	u32 reserved4;
+	u32 reserved5;
+	u16 reserved6;
+	u16 pec;			/* PCI event code */
+} __packed;
+
+static void zpci_event_log_err(struct zpci_ccdf_err *ccdf)
+{
+	struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid);
+
+	dev_err(&zdev->pdev->dev, "event code: 0x%x\n", ccdf->pec);
+}
+
+static void zpci_event_log_avail(struct zpci_ccdf_avail *ccdf)
+{
+	struct zpci_dev *zdev = get_zdev_by_fid(ccdf->fid);
+
+	pr_err("%s%s: availability event: fh: 0x%x  fid: 0x%x  event code: 0x%x  reason:",
+		(zdev) ? dev_driver_string(&zdev->pdev->dev) : "?",
+		(zdev) ? dev_name(&zdev->pdev->dev) : "?",
+		ccdf->fh, ccdf->fid, ccdf->pec);
+	print_hex_dump(KERN_CONT, "ccdf", DUMP_PREFIX_OFFSET,
+		       16, 1, ccdf, sizeof(*ccdf), false);
+
+	switch (ccdf->pec) {
+	case 0x0301:
+		zpci_enable_device(zdev);
+		break;
+	case 0x0302:
+		clp_add_pci_device(ccdf->fid, ccdf->fh, 0);
+		break;
+	case 0x0306:
+		clp_find_pci_devices();
+		break;
+	default:
+		break;
+	}
+}
+
+void zpci_event_error(void *data)
+{
+	struct zpci_ccdf_err *ccdf = data;
+	struct zpci_dev *zdev;
+
+	zpci_event_log_err(ccdf);
+	zdev = get_zdev_by_fid(ccdf->fid);
+	if (!zdev) {
+		pr_err("Error event for unknown fid: %x", ccdf->fid);
+		return;
+	}
+}
+
+void zpci_event_availability(void *data)
+{
+	zpci_event_log_avail(data);
+}
diff --git a/arch/s390/pci/pci_msi.c b/arch/s390/pci/pci_msi.c
new file mode 100644
index 000000000000..90fd3482b9e2
--- /dev/null
+++ b/arch/s390/pci/pci_msi.c
@@ -0,0 +1,141 @@
+/*
+ * Copyright IBM Corp. 2012
+ *
+ * Author(s):
+ *   Jan Glauber <jang@linux.vnet.ibm.com>
+ */
+
+#define COMPONENT "zPCI"
+#define pr_fmt(fmt) COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/err.h>
+#include <linux/rculist.h>
+#include <linux/hash.h>
+#include <linux/pci.h>
+#include <linux/msi.h>
+#include <asm/hw_irq.h>
+
+/* mapping of irq numbers to msi_desc */
+static struct hlist_head *msi_hash;
+static unsigned int msihash_shift = 6;
+#define msi_hashfn(nr)	hash_long(nr, msihash_shift)
+
+static DEFINE_SPINLOCK(msi_map_lock);
+
+struct msi_desc *__irq_get_msi_desc(unsigned int irq)
+{
+	struct hlist_node *entry;
+	struct msi_map *map;
+
+	hlist_for_each_entry_rcu(map, entry,
+			&msi_hash[msi_hashfn(irq)], msi_chain)
+		if (map->irq == irq)
+			return map->msi;
+	return NULL;
+}
+
+int zpci_msi_set_mask_bits(struct msi_desc *msi, u32 mask, u32 flag)
+{
+	if (msi->msi_attrib.is_msix) {
+		int offset = msi->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE +
+			PCI_MSIX_ENTRY_VECTOR_CTRL;
+		msi->masked = readl(msi->mask_base + offset);
+		writel(flag, msi->mask_base + offset);
+	} else {
+		if (msi->msi_attrib.maskbit) {
+			int pos;
+			u32 mask_bits;
+
+			pos = (long) msi->mask_base;
+			pci_read_config_dword(msi->dev, pos, &mask_bits);
+			mask_bits &= ~(mask);
+			mask_bits |= flag & mask;
+			pci_write_config_dword(msi->dev, pos, mask_bits);
+		} else {
+			return 0;
+		}
+	}
+
+	msi->msi_attrib.maskbit = !!flag;
+	return 1;
+}
+
+int zpci_setup_msi_irq(struct zpci_dev *zdev, struct msi_desc *msi,
+			unsigned int nr, int offset)
+{
+	struct msi_map *map;
+	struct msi_msg msg;
+	int rc;
+
+	map = kmalloc(sizeof(*map), GFP_KERNEL);
+	if (map == NULL)
+		return -ENOMEM;
+
+	map->irq = nr;
+	map->msi = msi;
+	zdev->msi_map[nr & ZPCI_MSI_MASK] = map;
+
+	pr_debug("%s hashing irq: %u  to bucket nr: %llu\n",
+		__func__, nr, msi_hashfn(nr));
+	hlist_add_head_rcu(&map->msi_chain, &msi_hash[msi_hashfn(nr)]);
+
+	spin_lock(&msi_map_lock);
+	rc = irq_set_msi_desc(nr, msi);
+	if (rc) {
+		spin_unlock(&msi_map_lock);
+		hlist_del_rcu(&map->msi_chain);
+		kfree(map);
+		zdev->msi_map[nr & ZPCI_MSI_MASK] = NULL;
+		return rc;
+	}
+	spin_unlock(&msi_map_lock);
+
+	msg.data = nr - offset;
+	msg.address_lo = zdev->msi_addr & 0xffffffff;
+	msg.address_hi = zdev->msi_addr >> 32;
+	write_msi_msg(nr, &msg);
+	return 0;
+}
+
+void zpci_teardown_msi_irq(struct zpci_dev *zdev, struct msi_desc *msi)
+{
+	int irq = msi->irq & ZPCI_MSI_MASK;
+	struct msi_map *map;
+
+	msi->msg.address_lo = 0;
+	msi->msg.address_hi = 0;
+	msi->msg.data = 0;
+	msi->irq = 0;
+	zpci_msi_set_mask_bits(msi, 1, 1);
+
+	spin_lock(&msi_map_lock);
+	map = zdev->msi_map[irq];
+	hlist_del_rcu(&map->msi_chain);
+	kfree(map);
+	zdev->msi_map[irq] = NULL;
+	spin_unlock(&msi_map_lock);
+}
+
+/*
+ * The msi hash table has 256 entries which is good for 4..20
+ * devices (a typical device allocates 10 + CPUs MSI's). Maybe make
+ * the hash table size adjustable later.
+ */
+int __init zpci_msihash_init(void)
+{
+	unsigned int i;
+
+	msi_hash = kmalloc(256 * sizeof(*msi_hash), GFP_KERNEL);
+	if (!msi_hash)
+		return -ENOMEM;
+
+	for (i = 0; i < (1U << msihash_shift); i++)
+		INIT_HLIST_HEAD(&msi_hash[i]);
+	return 0;
+}
+
+void __init zpci_msihash_exit(void)
+{
+	kfree(msi_hash);
+}
diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c
new file mode 100644
index 000000000000..a42cce69d0a0
--- /dev/null
+++ b/arch/s390/pci/pci_sysfs.c
@@ -0,0 +1,86 @@
+/*
+ * Copyright IBM Corp. 2012
+ *
+ * Author(s):
+ *   Jan Glauber <jang@linux.vnet.ibm.com>
+ */
+
+#define COMPONENT "zPCI"
+#define pr_fmt(fmt) COMPONENT ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/stat.h>
+#include <linux/pci.h>
+
+static ssize_t show_fid(struct device *dev, struct device_attribute *attr,
+			char *buf)
+{
+	struct zpci_dev *zdev = get_zdev(container_of(dev, struct pci_dev, dev));
+
+	sprintf(buf, "0x%08x\n", zdev->fid);
+	return strlen(buf);
+}
+static DEVICE_ATTR(function_id, S_IRUGO, show_fid, NULL);
+
+static ssize_t show_fh(struct device *dev, struct device_attribute *attr,
+		       char *buf)
+{
+	struct zpci_dev *zdev = get_zdev(container_of(dev, struct pci_dev, dev));
+
+	sprintf(buf, "0x%08x\n", zdev->fh);
+	return strlen(buf);
+}
+static DEVICE_ATTR(function_handle, S_IRUGO, show_fh, NULL);
+
+static ssize_t show_pchid(struct device *dev, struct device_attribute *attr,
+			  char *buf)
+{
+	struct zpci_dev *zdev = get_zdev(container_of(dev, struct pci_dev, dev));
+
+	sprintf(buf, "0x%04x\n", zdev->pchid);
+	return strlen(buf);
+}
+static DEVICE_ATTR(pchid, S_IRUGO, show_pchid, NULL);
+
+static ssize_t show_pfgid(struct device *dev, struct device_attribute *attr,
+			  char *buf)
+{
+	struct zpci_dev *zdev = get_zdev(container_of(dev, struct pci_dev, dev));
+
+	sprintf(buf, "0x%02x\n", zdev->pfgid);
+	return strlen(buf);
+}
+static DEVICE_ATTR(pfgid, S_IRUGO, show_pfgid, NULL);
+
+static struct device_attribute *zpci_dev_attrs[] = {
+	&dev_attr_function_id,
+	&dev_attr_function_handle,
+	&dev_attr_pchid,
+	&dev_attr_pfgid,
+	NULL,
+};
+
+int zpci_sysfs_add_device(struct device *dev)
+{
+	int i, rc = 0;
+
+	for (i = 0; zpci_dev_attrs[i]; i++) {
+		rc = device_create_file(dev, zpci_dev_attrs[i]);
+		if (rc)
+			goto error;
+	}
+	return 0;
+
+error:
+	while (--i >= 0)
+		device_remove_file(dev, zpci_dev_attrs[i]);
+	return rc;
+}
+
+void zpci_sysfs_remove_device(struct device *dev)
+{
+	int i;
+
+	for (i = 0; zpci_dev_attrs[i]; i++)
+		device_remove_file(dev, zpci_dev_attrs[i]);
+}