linux_kernel/arch/powerpc/platforms/powernv/pci.c

/*
 * Support PCI/PCIe on PowerNV platforms
 *
 * Currently supports only P5IOC2
 *
 * Copyright 2011 Benjamin Herrenschmidt, IBM Corp.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
 * as published by the Free Software Foundation; either version
 * 2 of the License, or (at your option) any later version.
 */

#include <linux/kernel.h>
#include <linux/pci.h>
#include <linux/delay.h>
#include <linux/string.h>
#include <linux/init.h>
#include <linux/irq.h>
#include <linux/io.h>
#include <linux/msi.h>
#include <linux/iommu.h>

#include <asm/sections.h>
#include <asm/io.h>
#include <asm/prom.h>
#include <asm/pci-bridge.h>
#include <asm/machdep.h>
#include <asm/msi_bitmap.h>
#include <asm/ppc-pci.h>
#include <asm/opal.h>
#include <asm/iommu.h>
#include <asm/tce.h>
#include <asm/firmware.h>
#include <asm/eeh_event.h>
#include <asm/eeh.h>

#include "powernv.h"
#include "pci.h"

/* Delay in usec */
#define PCI_RESET_DELAY_US	3000000

#define cfg_dbg(fmt...)	do { } while(0)
//#define cfg_dbg(fmt...)	printk(fmt)

#ifdef CONFIG_PCI_MSI
int pnv_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type)
{
	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
	struct pnv_phb *phb = hose->private_data;
	struct msi_desc *entry;
	struct msi_msg msg;
	int hwirq;
	unsigned int virq;
	int rc;

	if (WARN_ON(!phb) || !phb->msi_bmp.bitmap)
		return -ENODEV;

	if (pdev->no_64bit_msi && !phb->msi32_support)
		return -ENODEV;

	list_for_each_entry(entry, &pdev->msi_list, list) {
		if (!entry->msi_attrib.is_64 && !phb->msi32_support) {
			pr_warn("%s: Supports only 64-bit MSIs\n",
				pci_name(pdev));
			return -ENXIO;
		}
		hwirq = msi_bitmap_alloc_hwirqs(&phb->msi_bmp, 1);
		if (hwirq < 0) {
			pr_warn("%s: Failed to find a free MSI\n",
				pci_name(pdev));
			return -ENOSPC;
		}
		virq = irq_create_mapping(NULL, phb->msi_base + hwirq);
		if (virq == NO_IRQ) {
			pr_warn("%s: Failed to map MSI to linux irq\n",
				pci_name(pdev));
			msi_bitmap_free_hwirqs(&phb->msi_bmp, hwirq, 1);
			return -ENOMEM;
		}
		rc = phb->msi_setup(phb, pdev, phb->msi_base + hwirq,
				    virq, entry->msi_attrib.is_64, &msg);
		if (rc) {
			pr_warn("%s: Failed to setup MSI\n", pci_name(pdev));
			irq_dispose_mapping(virq);
			msi_bitmap_free_hwirqs(&phb->msi_bmp, hwirq, 1);
			return rc;
		}
		irq_set_msi_desc(virq, entry);
		pci_write_msi_msg(virq, &msg);
	}
	return 0;
}

void pnv_teardown_msi_irqs(struct pci_dev *pdev)
{
	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
	struct pnv_phb *phb = hose->private_data;
	struct msi_desc *entry;

	if (WARN_ON(!phb))
		return;

	list_for_each_entry(entry, &pdev->msi_list, list) {
		if (entry->irq == NO_IRQ)
			continue;
		irq_set_msi_desc(entry->irq, NULL);
		msi_bitmap_free_hwirqs(&phb->msi_bmp,
			virq_to_hw(entry->irq) - phb->msi_base, 1);
		irq_dispose_mapping(entry->irq);
	}
}
#endif /* CONFIG_PCI_MSI */

static void pnv_pci_dump_p7ioc_diag_data(struct pci_controller *hose,
					 struct OpalIoPhbErrorCommon *common)
{
	struct OpalIoP7IOCPhbErrorData *data;
	int i;

	data = (struct OpalIoP7IOCPhbErrorData *)common;
	pr_info("P7IOC PHB#%d Diag-data (Version: %d)\n",
		hose->global_number, be32_to_cpu(common->version));

	if (data->brdgCtl)
		pr_info("brdgCtl:     %08x\n",
			be32_to_cpu(data->brdgCtl));
	if (data->portStatusReg || data->rootCmplxStatus ||
	    data->busAgentStatus)
		pr_info("UtlSts:      %08x %08x %08x\n",
			be32_to_cpu(data->portStatusReg),
			be32_to_cpu(data->rootCmplxStatus),
			be32_to_cpu(data->busAgentStatus));
	if (data->deviceStatus || data->slotStatus   ||
	    data->linkStatus   || data->devCmdStatus ||
	    data->devSecStatus)
		pr_info("RootSts:     %08x %08x %08x %08x %08x\n",
			be32_to_cpu(data->deviceStatus),
			be32_to_cpu(data->slotStatus),
			be32_to_cpu(data->linkStatus),
			be32_to_cpu(data->devCmdStatus),
			be32_to_cpu(data->devSecStatus));
	if (data->rootErrorStatus   || data->uncorrErrorStatus ||
	    data->corrErrorStatus)
		pr_info("RootErrSts:  %08x %08x %08x\n",
			be32_to_cpu(data->rootErrorStatus),
			be32_to_cpu(data->uncorrErrorStatus),
			be32_to_cpu(data->corrErrorStatus));
	if (data->tlpHdr1 || data->tlpHdr2 ||
	    data->tlpHdr3 || data->tlpHdr4)
		pr_info("RootErrLog:  %08x %08x %08x %08x\n",
			be32_to_cpu(data->tlpHdr1),
			be32_to_cpu(data->tlpHdr2),
			be32_to_cpu(data->tlpHdr3),
			be32_to_cpu(data->tlpHdr4));
	if (data->sourceId || data->errorClass ||
	    data->correlator)
		pr_info("RootErrLog1: %08x %016llx %016llx\n",
			be32_to_cpu(data->sourceId),
			be64_to_cpu(data->errorClass),
			be64_to_cpu(data->correlator));
	if (data->p7iocPlssr || data->p7iocCsr)
		pr_info("PhbSts:      %016llx %016llx\n",
			be64_to_cpu(data->p7iocPlssr),
			be64_to_cpu(data->p7iocCsr));
	if (data->lemFir)
		pr_info("Lem:         %016llx %016llx %016llx\n",
			be64_to_cpu(data->lemFir),
			be64_to_cpu(data->lemErrorMask),
			be64_to_cpu(data->lemWOF));
	if (data->phbErrorStatus)
		pr_info("PhbErr:      %016llx %016llx %016llx %016llx\n",
			be64_to_cpu(data->phbErrorStatus),
			be64_to_cpu(data->phbFirstErrorStatus),
			be64_to_cpu(data->phbErrorLog0),
			be64_to_cpu(data->phbErrorLog1));
	if (data->mmioErrorStatus)
		pr_info("OutErr:      %016llx %016llx %016llx %016llx\n",
			be64_to_cpu(data->mmioErrorStatus),
			be64_to_cpu(data->mmioFirstErrorStatus),
			be64_to_cpu(data->mmioErrorLog0),
			be64_to_cpu(data->mmioErrorLog1));
	if (data->dma0ErrorStatus)
		pr_info("InAErr:      %016llx %016llx %016llx %016llx\n",
			be64_to_cpu(data->dma0ErrorStatus),
			be64_to_cpu(data->dma0FirstErrorStatus),
			be64_to_cpu(data->dma0ErrorLog0),
			be64_to_cpu(data->dma0ErrorLog1));
	if (data->dma1ErrorStatus)
		pr_info("InBErr:      %016llx %016llx %016llx %016llx\n",
			be64_to_cpu(data->dma1ErrorStatus),
			be64_to_cpu(data->dma1FirstErrorStatus),
			be64_to_cpu(data->dma1ErrorLog0),
			be64_to_cpu(data->dma1ErrorLog1));

	for (i = 0; i < OPAL_P7IOC_NUM_PEST_REGS; i++) {
		if ((data->pestA[i] >> 63) == 0 &&
		    (data->pestB[i] >> 63) == 0)
			continue;

		pr_info("PE[%3d] A/B: %016llx %016llx\n",
			i, be64_to_cpu(data->pestA[i]),
			be64_to_cpu(data->pestB[i]));
	}
}

static void pnv_pci_dump_phb3_diag_data(struct pci_controller *hose,
					struct OpalIoPhbErrorCommon *common)
{
	struct OpalIoPhb3ErrorData *data;
	int i;

	data = (struct OpalIoPhb3ErrorData*)common;
	pr_info("PHB3 PHB#%d Diag-data (Version: %d)\n",
		hose->global_number, be32_to_cpu(common->version));
	if (data->brdgCtl)
		pr_info("brdgCtl:     %08x\n",
			be32_to_cpu(data->brdgCtl));
	if (data->portStatusReg || data->rootCmplxStatus ||
	    data->busAgentStatus)
		pr_info("UtlSts:      %08x %08x %08x\n",
			be32_to_cpu(data->portStatusReg),
			be32_to_cpu(data->rootCmplxStatus),
			be32_to_cpu(data->busAgentStatus));
	if (data->deviceStatus || data->slotStatus   ||
	    data->linkStatus   || data->devCmdStatus ||
	    data->devSecStatus)
		pr_info("RootSts:     %08x %08x %08x %08x %08x\n",
			be32_to_cpu(data->deviceStatus),
			be32_to_cpu(data->slotStatus),
			be32_to_cpu(data->linkStatus),
			be32_to_cpu(data->devCmdStatus),
			be32_to_cpu(data->devSecStatus));
	if (data->rootErrorStatus || data->uncorrErrorStatus ||
	    data->corrErrorStatus)
		pr_info("RootErrSts:  %08x %08x %08x\n",
			be32_to_cpu(data->rootErrorStatus),
			be32_to_cpu(data->uncorrErrorStatus),
			be32_to_cpu(data->corrErrorStatus));
	if (data->tlpHdr1 || data->tlpHdr2 ||
	    data->tlpHdr3 || data->tlpHdr4)
		pr_info("RootErrLog:  %08x %08x %08x %08x\n",
			be32_to_cpu(data->tlpHdr1),
			be32_to_cpu(data->tlpHdr2),
			be32_to_cpu(data->tlpHdr3),
			be32_to_cpu(data->tlpHdr4));
	if (data->sourceId || data->errorClass ||
	    data->correlator)
		pr_info("RootErrLog1: %08x %016llx %016llx\n",
			be32_to_cpu(data->sourceId),
			be64_to_cpu(data->errorClass),
			be64_to_cpu(data->correlator));
	if (data->nFir)
		pr_info("nFir:        %016llx %016llx %016llx\n",
			be64_to_cpu(data->nFir),
			be64_to_cpu(data->nFirMask),
			be64_to_cpu(data->nFirWOF));
	if (data->phbPlssr || data->phbCsr)
		pr_info("PhbSts:      %016llx %016llx\n",
			be64_to_cpu(data->phbPlssr),
			be64_to_cpu(data->phbCsr));
	if (data->lemFir)
		pr_info("Lem:         %016llx %016llx %016llx\n",
			be64_to_cpu(data->lemFir),
			be64_to_cpu(data->lemErrorMask),
			be64_to_cpu(data->lemWOF));
	if (data->phbErrorStatus)
		pr_info("PhbErr:      %016llx %016llx %016llx %016llx\n",
			be64_to_cpu(data->phbErrorStatus),
			be64_to_cpu(data->phbFirstErrorStatus),
			be64_to_cpu(data->phbErrorLog0),
			be64_to_cpu(data->phbErrorLog1));
	if (data->mmioErrorStatus)
		pr_info("OutErr:      %016llx %016llx %016llx %016llx\n",
			be64_to_cpu(data->mmioErrorStatus),
			be64_to_cpu(data->mmioFirstErrorStatus),
			be64_to_cpu(data->mmioErrorLog0),
			be64_to_cpu(data->mmioErrorLog1));
	if (data->dma0ErrorStatus)
		pr_info("InAErr:      %016llx %016llx %016llx %016llx\n",
			be64_to_cpu(data->dma0ErrorStatus),
			be64_to_cpu(data->dma0FirstErrorStatus),
			be64_to_cpu(data->dma0ErrorLog0),
			be64_to_cpu(data->dma0ErrorLog1));
	if (data->dma1ErrorStatus)
		pr_info("InBErr:      %016llx %016llx %016llx %016llx\n",
			be64_to_cpu(data->dma1ErrorStatus),
			be64_to_cpu(data->dma1FirstErrorStatus),
			be64_to_cpu(data->dma1ErrorLog0),
			be64_to_cpu(data->dma1ErrorLog1));

	for (i = 0; i < OPAL_PHB3_NUM_PEST_REGS; i++) {
		if ((be64_to_cpu(data->pestA[i]) >> 63) == 0 &&
		    (be64_to_cpu(data->pestB[i]) >> 63) == 0)
			continue;

		pr_info("PE[%3d] A/B: %016llx %016llx\n",
				i, be64_to_cpu(data->pestA[i]),
				be64_to_cpu(data->pestB[i]));
	}
}

void pnv_pci_dump_phb_diag_data(struct pci_controller *hose,
				unsigned char *log_buff)
{
	struct OpalIoPhbErrorCommon *common;

	if (!hose || !log_buff)
		return;

	common = (struct OpalIoPhbErrorCommon *)log_buff;
	switch (be32_to_cpu(common->ioType)) {
	case OPAL_PHB_ERROR_DATA_TYPE_P7IOC:
		pnv_pci_dump_p7ioc_diag_data(hose, common);
		break;
	case OPAL_PHB_ERROR_DATA_TYPE_PHB3:
		pnv_pci_dump_phb3_diag_data(hose, common);
		break;
	default:
		pr_warn("%s: Unrecognized ioType %d\n",
			__func__, be32_to_cpu(common->ioType));
	}
}

static void pnv_pci_handle_eeh_config(struct pnv_phb *phb, u32 pe_no)
{
	unsigned long flags, rc;
	int has_diag, ret = 0;

	spin_lock_irqsave(&phb->lock, flags);

	/* Fetch PHB diag-data */
	rc = opal_pci_get_phb_diag_data2(phb->opal_id, phb->diag.blob,
					 PNV_PCI_DIAG_BUF_SIZE);
	has_diag = (rc == OPAL_SUCCESS);

	/* If PHB supports compound PE, to handle it */
	if (phb->unfreeze_pe) {
		ret = phb->unfreeze_pe(phb,
				       pe_no,
				       OPAL_EEH_ACTION_CLEAR_FREEZE_ALL);
	} else {
		rc = opal_pci_eeh_freeze_clear(phb->opal_id,
					     pe_no,
					     OPAL_EEH_ACTION_CLEAR_FREEZE_ALL);
		if (rc) {
			pr_warn("%s: Failure %ld clearing frozen "
				"PHB#%x-PE#%x\n",
				__func__, rc, phb->hose->global_number,
				pe_no);
			ret = -EIO;
		}
	}

	/*
	 * For now, let's only display the diag buffer when we fail to clear
	 * the EEH status. We'll do more sensible things later when we have
	 * proper EEH support. We need to make sure we don't pollute ourselves
	 * with the normal errors generated when probing empty slots
	 */
	if (has_diag && ret)
		pnv_pci_dump_phb_diag_data(phb->hose, phb->diag.blob);

	spin_unlock_irqrestore(&phb->lock, flags);
}

static void pnv_pci_config_check_eeh(struct pci_dn *pdn)
{
	struct pnv_phb *phb = pdn->phb->private_data;
	u8	fstate;
	__be16	pcierr;
	int	pe_no;
	s64	rc;

	/*
	 * Get the PE#. During the PCI probe stage, we might not
	 * setup that yet. So all ER errors should be mapped to
	 * reserved PE.
	 */
	pe_no = pdn->pe_number;
	if (pe_no == IODA_INVALID_PE) {
		if (phb->type == PNV_PHB_P5IOC2)
			pe_no = 0;
		else
			pe_no = phb->ioda.reserved_pe;
	}

	/*
	 * Fetch frozen state. If the PHB support compound PE,
	 * we need handle that case.
	 */
	if (phb->get_pe_state) {
		fstate = phb->get_pe_state(phb, pe_no);
	} else {
		rc = opal_pci_eeh_freeze_status(phb->opal_id,
						pe_no,
						&fstate,
						&pcierr,
						NULL);
		if (rc) {
			pr_warn("%s: Failure %lld getting PHB#%x-PE#%x state\n",
				__func__, rc, phb->hose->global_number, pe_no);
			return;
		}
	}

	cfg_dbg(" -> EEH check, bdfn=%04x PE#%d fstate=%x\n",
		(pdn->busno << 8) | (pdn->devfn), pe_no, fstate);

	/* Clear the frozen state if applicable */
	if (fstate == OPAL_EEH_STOPPED_MMIO_FREEZE ||
	    fstate == OPAL_EEH_STOPPED_DMA_FREEZE  ||
	    fstate == OPAL_EEH_STOPPED_MMIO_DMA_FREEZE) {
		/*
		 * If PHB supports compound PE, freeze it for
		 * consistency.
		 */
		if (phb->freeze_pe)
			phb->freeze_pe(phb, pe_no);

		pnv_pci_handle_eeh_config(phb, pe_no);
	}
}

int pnv_pci_cfg_read(struct pci_dn *pdn,
		     int where, int size, u32 *val)
{
	struct pnv_phb *phb = pdn->phb->private_data;
	u32 bdfn = (pdn->busno << 8) | pdn->devfn;
	s64 rc;

	switch (size) {
	case 1: {
		u8 v8;
		rc = opal_pci_config_read_byte(phb->opal_id, bdfn, where, &v8);
		*val = (rc == OPAL_SUCCESS) ? v8 : 0xff;
		break;
	}
	case 2: {
		__be16 v16;
		rc = opal_pci_config_read_half_word(phb->opal_id, bdfn, where,
						   &v16);
		*val = (rc == OPAL_SUCCESS) ? be16_to_cpu(v16) : 0xffff;
		break;
	}
	case 4: {
		__be32 v32;
		rc = opal_pci_config_read_word(phb->opal_id, bdfn, where, &v32);
		*val = (rc == OPAL_SUCCESS) ? be32_to_cpu(v32) : 0xffffffff;
		break;
	}
	default:
		return PCIBIOS_FUNC_NOT_SUPPORTED;
	}

	cfg_dbg("%s: bus: %x devfn: %x +%x/%x -> %08x\n",
		__func__, pdn->busno, pdn->devfn, where, size, *val);
	return PCIBIOS_SUCCESSFUL;
}

int pnv_pci_cfg_write(struct pci_dn *pdn,
		      int where, int size, u32 val)
{
	struct pnv_phb *phb = pdn->phb->private_data;
	u32 bdfn = (pdn->busno << 8) | pdn->devfn;

	cfg_dbg("%s: bus: %x devfn: %x +%x/%x -> %08x\n",
		pdn->busno, pdn->devfn, where, size, val);
	switch (size) {
	case 1:
		opal_pci_config_write_byte(phb->opal_id, bdfn, where, val);
		break;
	case 2:
		opal_pci_config_write_half_word(phb->opal_id, bdfn, where, val);
		break;
	case 4:
		opal_pci_config_write_word(phb->opal_id, bdfn, where, val);
		break;
	default:
		return PCIBIOS_FUNC_NOT_SUPPORTED;
	}

	return PCIBIOS_SUCCESSFUL;
}

#if CONFIG_EEH
static bool pnv_pci_cfg_check(struct pci_dn *pdn)
{
	struct eeh_dev *edev = NULL;
	struct pnv_phb *phb = pdn->phb->private_data;

	/* EEH not enabled ? */
	if (!(phb->flags & PNV_PHB_FLAG_EEH))
		return true;

	/* PE reset or device removed ? */
	edev = pdn->edev;
	if (edev) {
		if (edev->pe &&
		    (edev->pe->state & EEH_PE_CFG_BLOCKED))
			return false;

		if (edev->mode & EEH_DEV_REMOVED)
			return false;
	}

	return true;
}
#else
static inline pnv_pci_cfg_check(struct pci_dn *pdn)
{
	return true;
}
#endif /* CONFIG_EEH */

static int pnv_pci_read_config(struct pci_bus *bus,
			       unsigned int devfn,
			       int where, int size, u32 *val)
{
	struct pci_dn *pdn;
	struct pnv_phb *phb;
	int ret;

	*val = 0xFFFFFFFF;
	pdn = pci_get_pdn_by_devfn(bus, devfn);
	if (!pdn)
		return PCIBIOS_DEVICE_NOT_FOUND;

	if (!pnv_pci_cfg_check(pdn))
		return PCIBIOS_DEVICE_NOT_FOUND;

	ret = pnv_pci_cfg_read(pdn, where, size, val);
	phb = pdn->phb->private_data;
	if (phb->flags & PNV_PHB_FLAG_EEH && pdn->edev) {
		if (*val == EEH_IO_ERROR_VALUE(size) &&
		    eeh_dev_check_failure(pdn->edev))
                        return PCIBIOS_DEVICE_NOT_FOUND;
	} else {
		pnv_pci_config_check_eeh(pdn);
	}

	return ret;
}

static int pnv_pci_write_config(struct pci_bus *bus,
				unsigned int devfn,
				int where, int size, u32 val)
{
	struct pci_dn *pdn;
	struct pnv_phb *phb;
	int ret;

	pdn = pci_get_pdn_by_devfn(bus, devfn);
	if (!pdn)
		return PCIBIOS_DEVICE_NOT_FOUND;

	if (!pnv_pci_cfg_check(pdn))
		return PCIBIOS_DEVICE_NOT_FOUND;

	ret = pnv_pci_cfg_write(pdn, where, size, val);
	phb = pdn->phb->private_data;
	if (!(phb->flags & PNV_PHB_FLAG_EEH))
		pnv_pci_config_check_eeh(pdn);

	return ret;
}

struct pci_ops pnv_pci_ops = {
	.read  = pnv_pci_read_config,
	.write = pnv_pci_write_config,
};

static __be64 *pnv_tce(struct iommu_table *tbl, long idx)
{
	__be64 *tmp = ((__be64 *)tbl->it_base);
	int  level = tbl->it_indirect_levels;
	const long shift = ilog2(tbl->it_level_size);
	unsigned long mask = (tbl->it_level_size - 1) << (level * shift);

	while (level) {
		int n = (idx & mask) >> (level * shift);
		unsigned long tce = be64_to_cpu(tmp[n]);

		tmp = __va(tce & ~(TCE_PCI_READ | TCE_PCI_WRITE));
		idx &= ~mask;
		mask >>= shift;
		--level;
	}

	return tmp + idx;
}

int pnv_tce_build(struct iommu_table *tbl, long index, long npages,
		unsigned long uaddr, enum dma_data_direction direction,
		struct dma_attrs *attrs)
{
	u64 proto_tce = iommu_direction_to_tce_perm(direction);
	u64 rpn = __pa(uaddr) >> tbl->it_page_shift;
	long i;

	for (i = 0; i < npages; i++) {
		unsigned long newtce = proto_tce |
			((rpn + i) << tbl->it_page_shift);
		unsigned long idx = index - tbl->it_offset + i;

		*(pnv_tce(tbl, idx)) = cpu_to_be64(newtce);
	}

	return 0;
}

#ifdef CONFIG_IOMMU_API
int pnv_tce_xchg(struct iommu_table *tbl, long index,
		unsigned long *hpa, enum dma_data_direction *direction)
{
	u64 proto_tce = iommu_direction_to_tce_perm(*direction);
	unsigned long newtce = *hpa | proto_tce, oldtce;
	unsigned long idx = index - tbl->it_offset;

	BUG_ON(*hpa & ~IOMMU_PAGE_MASK(tbl));

	oldtce = xchg(pnv_tce(tbl, idx), cpu_to_be64(newtce));
	*hpa = be64_to_cpu(oldtce) & ~(TCE_PCI_READ | TCE_PCI_WRITE);
	*direction = iommu_tce_direction(oldtce);

	return 0;
}
#endif

void pnv_tce_free(struct iommu_table *tbl, long index, long npages)
{
	long i;

	for (i = 0; i < npages; i++) {
		unsigned long idx = index - tbl->it_offset + i;

		*(pnv_tce(tbl, idx)) = cpu_to_be64(0);
	}
}

unsigned long pnv_tce_get(struct iommu_table *tbl, long index)
{
	return *(pnv_tce(tbl, index - tbl->it_offset));
}

struct iommu_table *pnv_pci_table_alloc(int nid)
{
	struct iommu_table *tbl;

	tbl = kzalloc_node(sizeof(struct iommu_table), GFP_KERNEL, nid);
	INIT_LIST_HEAD_RCU(&tbl->it_group_list);

	return tbl;
}

long pnv_pci_link_table_and_group(int node, int num,
		struct iommu_table *tbl,
		struct iommu_table_group *table_group)
{
	struct iommu_table_group_link *tgl = NULL;

	if (WARN_ON(!tbl || !table_group))
		return -EINVAL;

	tgl = kzalloc_node(sizeof(struct iommu_table_group_link), GFP_KERNEL,
			node);
	if (!tgl)
		return -ENOMEM;

	tgl->table_group = table_group;
	list_add_rcu(&tgl->next, &tbl->it_group_list);

	table_group->tables[num] = tbl;

	return 0;
}

static void pnv_iommu_table_group_link_free(struct rcu_head *head)
{
	struct iommu_table_group_link *tgl = container_of(head,
			struct iommu_table_group_link, rcu);

	kfree(tgl);
}

void pnv_pci_unlink_table_and_group(struct iommu_table *tbl,
		struct iommu_table_group *table_group)
{
	long i;
	bool found;
	struct iommu_table_group_link *tgl;

	if (!tbl || !table_group)
		return;

	/* Remove link to a group from table's list of attached groups */
	found = false;
	list_for_each_entry_rcu(tgl, &tbl->it_group_list, next) {
		if (tgl->table_group == table_group) {
			list_del_rcu(&tgl->next);
			call_rcu(&tgl->rcu, pnv_iommu_table_group_link_free);
			found = true;
			break;
		}
	}
	if (WARN_ON(!found))
		return;

	/* Clean a pointer to iommu_table in iommu_table_group::tables[] */
	found = false;
	for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
		if (table_group->tables[i] == tbl) {
			table_group->tables[i] = NULL;
			found = true;
			break;
		}
	}
	WARN_ON(!found);
}

void pnv_pci_setup_iommu_table(struct iommu_table *tbl,
			       void *tce_mem, u64 tce_size,
			       u64 dma_offset, unsigned page_shift)
{
	tbl->it_blocksize = 16;
	tbl->it_base = (unsigned long)tce_mem;
	tbl->it_page_shift = page_shift;
	tbl->it_offset = dma_offset >> tbl->it_page_shift;
	tbl->it_index = 0;
	tbl->it_size = tce_size >> 3;
	tbl->it_busno = 0;
	tbl->it_type = TCE_PCI;
}

void pnv_pci_dma_dev_setup(struct pci_dev *pdev)
{
	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
	struct pnv_phb *phb = hose->private_data;
#ifdef CONFIG_PCI_IOV
	struct pnv_ioda_pe *pe;
	struct pci_dn *pdn;

	/* Fix the VF pdn PE number */
	if (pdev->is_virtfn) {
		pdn = pci_get_pdn(pdev);
		WARN_ON(pdn->pe_number != IODA_INVALID_PE);
		list_for_each_entry(pe, &phb->ioda.pe_list, list) {
			if (pe->rid == ((pdev->bus->number << 8) |
			    (pdev->devfn & 0xff))) {
				pdn->pe_number = pe->pe_number;
				pe->pdev = pdev;
				break;
			}
		}
	}
#endif /* CONFIG_PCI_IOV */

	if (phb && phb->dma_dev_setup)
		phb->dma_dev_setup(phb, pdev);
}

u64 pnv_pci_dma_get_required_mask(struct pci_dev *pdev)
{
	struct pci_controller *hose = pci_bus_to_host(pdev->bus);
	struct pnv_phb *phb = hose->private_data;

	if (phb && phb->dma_get_required_mask)
		return phb->dma_get_required_mask(phb, pdev);

	return __dma_get_required_mask(&pdev->dev);
}

void pnv_pci_shutdown(void)
{
	struct pci_controller *hose;

	list_for_each_entry(hose, &hose_list, list_node)
		if (hose->controller_ops.shutdown)
			hose->controller_ops.shutdown(hose);
}

/* Fixup wrong class code in p7ioc and p8 root complex */
static void pnv_p7ioc_rc_quirk(struct pci_dev *dev)
{
	dev->class = PCI_CLASS_BRIDGE_PCI << 8;
}
DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_IBM, 0x3b9, pnv_p7ioc_rc_quirk);

void __init pnv_pci_init(void)
{
	struct device_node *np;
	bool found_ioda = false;

	pci_add_flags(PCI_CAN_SKIP_ISA_ALIGN);

	/* If we don't have OPAL, eg. in sim, just skip PCI probe */
	if (!firmware_has_feature(FW_FEATURE_OPAL))
		return;

	/* Look for IODA IO-Hubs. We don't support mixing IODA
	 * and p5ioc2 due to the need to change some global
	 * probing flags
	 */
	for_each_compatible_node(np, NULL, "ibm,ioda-hub") {
		pnv_pci_init_ioda_hub(np);
		found_ioda = true;
	}

	/* Look for p5ioc2 IO-Hubs */
	if (!found_ioda)
		for_each_compatible_node(np, NULL, "ibm,p5ioc2")
			pnv_pci_init_p5ioc2_hub(np);

	/* Look for ioda2 built-in PHB3's */
	for_each_compatible_node(np, NULL, "ibm,ioda2-phb")
		pnv_pci_init_ioda2_phb(np);

	/* Setup the linkage between OF nodes and PHBs */
	pci_devs_phb_init();

	/* Configure IOMMU DMA hooks */
	set_pci_dma_ops(&dma_iommu_ops);
}

machine_subsys_initcall_sync(powernv, tce_iommu_bus_notifier_init);