/*
 *	linux/arch/alpha/kernel/sys_dp264.c
 *
 *	Copyright (C) 1995 David A Rusling
 *	Copyright (C) 1996, 1999 Jay A Estabrook
 *	Copyright (C) 1998, 1999 Richard Henderson
 *
 *	Modified by Christopher C. Chimelis, 2001 to
 *	add support for the addition of Shark to the
 *	Tsunami family.
 *
 * Code supporting the DP264 (EV6+TSUNAMI).
 */

#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/mm.h>
#include <linux/sched.h>
#include <linux/pci.h>
#include <linux/init.h>
#include <linux/bitops.h>

#include <asm/ptrace.h>
#include <asm/dma.h>
#include <asm/irq.h>
#include <asm/mmu_context.h>
#include <asm/io.h>
#include <asm/pgtable.h>
#include <asm/core_tsunami.h>
#include <asm/hwrpb.h>
#include <asm/tlbflush.h>

#include "proto.h"
#include "irq_impl.h"
#include "pci_impl.h"
#include "machvec_impl.h"


/* Note mask bit is true for ENABLED irqs.  */
static unsigned long cached_irq_mask;
/* dp264 boards handle at max four CPUs */
static unsigned long cpu_irq_affinity[4] = { 0UL, 0UL, 0UL, 0UL };

DEFINE_SPINLOCK(dp264_irq_lock);

static void
tsunami_update_irq_hw(unsigned long mask)
{
	register tsunami_cchip *cchip = TSUNAMI_cchip;
	unsigned long isa_enable = 1UL << 55;
	register int bcpu = boot_cpuid;

#ifdef CONFIG_SMP
	volatile unsigned long *dim0, *dim1, *dim2, *dim3;
	unsigned long mask0, mask1, mask2, mask3, dummy;

	mask &= ~isa_enable;
	mask0 = mask & cpu_irq_affinity[0];
	mask1 = mask & cpu_irq_affinity[1];
	mask2 = mask & cpu_irq_affinity[2];
	mask3 = mask & cpu_irq_affinity[3];

	if (bcpu == 0) mask0 |= isa_enable;
	else if (bcpu == 1) mask1 |= isa_enable;
	else if (bcpu == 2) mask2 |= isa_enable;
	else mask3 |= isa_enable;

	dim0 = &cchip->dim0.csr;
	dim1 = &cchip->dim1.csr;
	dim2 = &cchip->dim2.csr;
	dim3 = &cchip->dim3.csr;
	if (!cpu_possible(0)) dim0 = &dummy;
	if (!cpu_possible(1)) dim1 = &dummy;
	if (!cpu_possible(2)) dim2 = &dummy;
	if (!cpu_possible(3)) dim3 = &dummy;

	*dim0 = mask0;
	*dim1 = mask1;
	*dim2 = mask2;
	*dim3 = mask3;
	mb();
	*dim0;
	*dim1;
	*dim2;
	*dim3;
#else
	volatile unsigned long *dimB;
	if (bcpu == 0) dimB = &cchip->dim0.csr;
	else if (bcpu == 1) dimB = &cchip->dim1.csr;
	else if (bcpu == 2) dimB = &cchip->dim2.csr;
	else dimB = &cchip->dim3.csr;

	*dimB = mask | isa_enable;
	mb();
	*dimB;
#endif
}

static void
dp264_enable_irq(struct irq_data *d)
{
	spin_lock(&dp264_irq_lock);
	cached_irq_mask |= 1UL << d->irq;
	tsunami_update_irq_hw(cached_irq_mask);
	spin_unlock(&dp264_irq_lock);
}

static void
dp264_disable_irq(struct irq_data *d)
{
	spin_lock(&dp264_irq_lock);
	cached_irq_mask &= ~(1UL << d->irq);
	tsunami_update_irq_hw(cached_irq_mask);
	spin_unlock(&dp264_irq_lock);
}

static void
clipper_enable_irq(struct irq_data *d)
{
	spin_lock(&dp264_irq_lock);
	cached_irq_mask |= 1UL << (d->irq - 16);
	tsunami_update_irq_hw(cached_irq_mask);
	spin_unlock(&dp264_irq_lock);
}

static void
clipper_disable_irq(struct irq_data *d)
{
	spin_lock(&dp264_irq_lock);
	cached_irq_mask &= ~(1UL << (d->irq - 16));
	tsunami_update_irq_hw(cached_irq_mask);
	spin_unlock(&dp264_irq_lock);
}

static void
cpu_set_irq_affinity(unsigned int irq, cpumask_t affinity)
{
	int cpu;

	for (cpu = 0; cpu < 4; cpu++) {
		unsigned long aff = cpu_irq_affinity[cpu];
		if (cpumask_test_cpu(cpu, &affinity))
			aff |= 1UL << irq;
		else
			aff &= ~(1UL << irq);
		cpu_irq_affinity[cpu] = aff;
	}
}

static int
dp264_set_affinity(struct irq_data *d, const struct cpumask *affinity,
		   bool force)
{
	spin_lock(&dp264_irq_lock);
	cpu_set_irq_affinity(d->irq, *affinity);
	tsunami_update_irq_hw(cached_irq_mask);
	spin_unlock(&dp264_irq_lock);

	return 0;
}

static int
clipper_set_affinity(struct irq_data *d, const struct cpumask *affinity,
		     bool force)
{
	spin_lock(&dp264_irq_lock);
	cpu_set_irq_affinity(d->irq - 16, *affinity);
	tsunami_update_irq_hw(cached_irq_mask);
	spin_unlock(&dp264_irq_lock);

	return 0;
}

static struct irq_chip dp264_irq_type = {
	.name			= "DP264",
	.irq_unmask		= dp264_enable_irq,
	.irq_mask		= dp264_disable_irq,
	.irq_mask_ack		= dp264_disable_irq,
	.irq_set_affinity	= dp264_set_affinity,
};

static struct irq_chip clipper_irq_type = {
	.name			= "CLIPPER",
	.irq_unmask		= clipper_enable_irq,
	.irq_mask		= clipper_disable_irq,
	.irq_mask_ack		= clipper_disable_irq,
	.irq_set_affinity	= clipper_set_affinity,
};

static void
dp264_device_interrupt(unsigned long vector)
{
#if 1
	printk("dp264_device_interrupt: NOT IMPLEMENTED YET!!\n");
#else
	unsigned long pld;
	unsigned int i;

	/* Read the interrupt summary register of TSUNAMI */
	pld = TSUNAMI_cchip->dir0.csr;

	/*
	 * Now for every possible bit set, work through them and call
	 * the appropriate interrupt handler.
	 */
	while (pld) {
		i = ffz(~pld);
		pld &= pld - 1; /* clear least bit set */
		if (i == 55)
			isa_device_interrupt(vector);
		else
			handle_irq(16 + i);
#if 0
		TSUNAMI_cchip->dir0.csr = 1UL << i; mb();
		tmp = TSUNAMI_cchip->dir0.csr;
#endif
	}
#endif
}

static void 
dp264_srm_device_interrupt(unsigned long vector)
{
	int irq;

	irq = (vector - 0x800) >> 4;

	/*
	 * The SRM console reports PCI interrupts with a vector calculated by:
	 *
	 *	0x900 + (0x10 * DRIR-bit)
	 *
	 * So bit 16 shows up as IRQ 32, etc.
	 * 
	 * On DP264/BRICK/MONET, we adjust it down by 16 because at least
	 * that many of the low order bits of the DRIR are not used, and
	 * so we don't count them.
	 */
	if (irq >= 32)
		irq -= 16;

	handle_irq(irq);
}

static void 
clipper_srm_device_interrupt(unsigned long vector)
{
	int irq;

	irq = (vector - 0x800) >> 4;

/*
	 * The SRM console reports PCI interrupts with a vector calculated by:
	 *
	 *	0x900 + (0x10 * DRIR-bit)
	 *
	 * So bit 16 shows up as IRQ 32, etc.
	 * 
	 * CLIPPER uses bits 8-47 for PCI interrupts, so we do not need
	 * to scale down the vector reported, we just use it.
	 *
	 * Eg IRQ 24 is DRIR bit 8, etc, etc
	 */
	handle_irq(irq);
}

static void __init
init_tsunami_irqs(struct irq_chip * ops, int imin, int imax)
{
	long i;
	for (i = imin; i <= imax; ++i) {
		irq_set_chip_and_handler(i, ops, handle_level_irq);
		irq_set_status_flags(i, IRQ_LEVEL);
	}
}

static void __init
dp264_init_irq(void)
{
	outb(0, DMA1_RESET_REG);
	outb(0, DMA2_RESET_REG);
	outb(DMA_MODE_CASCADE, DMA2_MODE_REG);
	outb(0, DMA2_MASK_REG);

	if (alpha_using_srm)
		alpha_mv.device_interrupt = dp264_srm_device_interrupt;

	tsunami_update_irq_hw(0);

	init_i8259a_irqs();
	init_tsunami_irqs(&dp264_irq_type, 16, 47);
}

static void __init
clipper_init_irq(void)
{
	outb(0, DMA1_RESET_REG);
	outb(0, DMA2_RESET_REG);
	outb(DMA_MODE_CASCADE, DMA2_MODE_REG);
	outb(0, DMA2_MASK_REG);

	if (alpha_using_srm)
		alpha_mv.device_interrupt = clipper_srm_device_interrupt;

	tsunami_update_irq_hw(0);

	init_i8259a_irqs();
	init_tsunami_irqs(&clipper_irq_type, 24, 63);
}


/*
 * PCI Fixup configuration.
 *
 * Summary @ TSUNAMI_CSR_DIM0:
 * Bit      Meaning
 * 0-17     Unused
 *18        Interrupt SCSI B (Adaptec 7895 builtin)
 *19        Interrupt SCSI A (Adaptec 7895 builtin)
 *20        Interrupt Line D from slot 2 PCI0
 *21        Interrupt Line C from slot 2 PCI0
 *22        Interrupt Line B from slot 2 PCI0
 *23        Interrupt Line A from slot 2 PCI0
 *24        Interrupt Line D from slot 1 PCI0
 *25        Interrupt Line C from slot 1 PCI0
 *26        Interrupt Line B from slot 1 PCI0
 *27        Interrupt Line A from slot 1 PCI0
 *28        Interrupt Line D from slot 0 PCI0
 *29        Interrupt Line C from slot 0 PCI0
 *30        Interrupt Line B from slot 0 PCI0
 *31        Interrupt Line A from slot 0 PCI0
 *
 *32        Interrupt Line D from slot 3 PCI1
 *33        Interrupt Line C from slot 3 PCI1
 *34        Interrupt Line B from slot 3 PCI1
 *35        Interrupt Line A from slot 3 PCI1
 *36        Interrupt Line D from slot 2 PCI1
 *37        Interrupt Line C from slot 2 PCI1
 *38        Interrupt Line B from slot 2 PCI1
 *39        Interrupt Line A from slot 2 PCI1
 *40        Interrupt Line D from slot 1 PCI1
 *41        Interrupt Line C from slot 1 PCI1
 *42        Interrupt Line B from slot 1 PCI1
 *43        Interrupt Line A from slot 1 PCI1
 *44        Interrupt Line D from slot 0 PCI1
 *45        Interrupt Line C from slot 0 PCI1
 *46        Interrupt Line B from slot 0 PCI1
 *47        Interrupt Line A from slot 0 PCI1
 *48-52     Unused
 *53        PCI0 NMI (from Cypress)
 *54        PCI0 SMI INT (from Cypress)
 *55        PCI0 ISA Interrupt (from Cypress)
 *56-60     Unused
 *61        PCI1 Bus Error
 *62        PCI0 Bus Error
 *63        Reserved
 *
 * IdSel	
 *   5	 Cypress Bridge I/O
 *   6	 SCSI Adaptec builtin
 *   7	 64 bit PCI option slot 0 (all busses)
 *   8	 64 bit PCI option slot 1 (all busses)
 *   9	 64 bit PCI option slot 2 (all busses)
 *  10	 64 bit PCI option slot 3 (not bus 0)
 */

static int __init
isa_irq_fixup(const struct pci_dev *dev, int irq)
{
	u8 irq8;

	if (irq > 0)
		return irq;

	/* This interrupt is routed via ISA bridge, so we'll
	   just have to trust whatever value the console might
	   have assigned.  */
	pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &irq8);

	return irq8 & 0xf;
}

static int __init
dp264_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
{
	static char irq_tab[6][5] __initdata = {
		/*INT    INTA   INTB   INTC   INTD */
		{    -1,    -1,    -1,    -1,    -1}, /* IdSel 5 ISA Bridge */
		{ 16+ 3, 16+ 3, 16+ 2, 16+ 2, 16+ 2}, /* IdSel 6 SCSI builtin*/
		{ 16+15, 16+15, 16+14, 16+13, 16+12}, /* IdSel 7 slot 0 */
		{ 16+11, 16+11, 16+10, 16+ 9, 16+ 8}, /* IdSel 8 slot 1 */
		{ 16+ 7, 16+ 7, 16+ 6, 16+ 5, 16+ 4}, /* IdSel 9 slot 2 */
		{ 16+ 3, 16+ 3, 16+ 2, 16+ 1, 16+ 0}  /* IdSel 10 slot 3 */
	};
	const long min_idsel = 5, max_idsel = 10, irqs_per_slot = 5;
	struct pci_controller *hose = dev->sysdata;
	int irq = COMMON_TABLE_LOOKUP;

	if (irq > 0)
		irq += 16 * hose->index;

	return isa_irq_fixup(dev, irq);
}

static int __init
monet_map_irq(const struct pci_dev *dev, u8 slot, u8 pin)
{
	static char irq_tab[13][5] __initdata = {
		/*INT    INTA   INTB   INTC   INTD */
		{    45,    45,    45,    45,    45}, /* IdSel 3 21143 PCI1 */