ubuntu-linux-kernel/drivers/vme/bridges/vme_ca91cx42.c

1934 lines
48 KiB
C

/*
* Support for the Tundra Universe I/II VME-PCI Bridge Chips
*
* Author: Martyn Welch <martyn.welch@ge.com>
* Copyright 2008 GE Intelligent Platforms Embedded Systems, Inc.
*
* Based on work by Tom Armistead and Ajit Prem
* Copyright 2004 Motorola Inc.
*
* Derived from ca91c042.c by Michael Wyrick
*
* 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/module.h>
#include <linux/mm.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/pci.h>
#include <linux/dma-mapping.h>
#include <linux/poll.h>
#include <linux/interrupt.h>
#include <linux/spinlock.h>
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/time.h>
#include <linux/io.h>
#include <linux/uaccess.h>
#include <linux/vme.h>
#include "../vme_bridge.h"
#include "vme_ca91cx42.h"
static int ca91cx42_probe(struct pci_dev *, const struct pci_device_id *);
static void ca91cx42_remove(struct pci_dev *);
/* Module parameters */
static int geoid;
static const char driver_name[] = "vme_ca91cx42";
static const struct pci_device_id ca91cx42_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_TUNDRA, PCI_DEVICE_ID_TUNDRA_CA91C142) },
{ },
};
MODULE_DEVICE_TABLE(pci, ca91cx42_ids);
static struct pci_driver ca91cx42_driver = {
.name = driver_name,
.id_table = ca91cx42_ids,
.probe = ca91cx42_probe,
.remove = ca91cx42_remove,
};
static u32 ca91cx42_DMA_irqhandler(struct ca91cx42_driver *bridge)
{
wake_up(&bridge->dma_queue);
return CA91CX42_LINT_DMA;
}
static u32 ca91cx42_LM_irqhandler(struct ca91cx42_driver *bridge, u32 stat)
{
int i;
u32 serviced = 0;
for (i = 0; i < 4; i++) {
if (stat & CA91CX42_LINT_LM[i]) {
/* We only enable interrupts if the callback is set */
bridge->lm_callback[i](bridge->lm_data[i]);
serviced |= CA91CX42_LINT_LM[i];
}
}
return serviced;
}
/* XXX This needs to be split into 4 queues */
static u32 ca91cx42_MB_irqhandler(struct ca91cx42_driver *bridge, int mbox_mask)
{
wake_up(&bridge->mbox_queue);
return CA91CX42_LINT_MBOX;
}
static u32 ca91cx42_IACK_irqhandler(struct ca91cx42_driver *bridge)
{
wake_up(&bridge->iack_queue);
return CA91CX42_LINT_SW_IACK;
}
static u32 ca91cx42_VERR_irqhandler(struct vme_bridge *ca91cx42_bridge)
{
int val;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
val = ioread32(bridge->base + DGCS);
if (!(val & 0x00000800)) {
dev_err(ca91cx42_bridge->parent, "ca91cx42_VERR_irqhandler DMA "
"Read Error DGCS=%08X\n", val);
}
return CA91CX42_LINT_VERR;
}
static u32 ca91cx42_LERR_irqhandler(struct vme_bridge *ca91cx42_bridge)
{
int val;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
val = ioread32(bridge->base + DGCS);
if (!(val & 0x00000800))
dev_err(ca91cx42_bridge->parent, "ca91cx42_LERR_irqhandler DMA "
"Read Error DGCS=%08X\n", val);
return CA91CX42_LINT_LERR;
}
static u32 ca91cx42_VIRQ_irqhandler(struct vme_bridge *ca91cx42_bridge,
int stat)
{
int vec, i, serviced = 0;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
for (i = 7; i > 0; i--) {
if (stat & (1 << i)) {
vec = ioread32(bridge->base +
CA91CX42_V_STATID[i]) & 0xff;
vme_irq_handler(ca91cx42_bridge, i, vec);
serviced |= (1 << i);
}
}
return serviced;
}
static irqreturn_t ca91cx42_irqhandler(int irq, void *ptr)
{
u32 stat, enable, serviced = 0;
struct vme_bridge *ca91cx42_bridge;
struct ca91cx42_driver *bridge;
ca91cx42_bridge = ptr;
bridge = ca91cx42_bridge->driver_priv;
enable = ioread32(bridge->base + LINT_EN);
stat = ioread32(bridge->base + LINT_STAT);
/* Only look at unmasked interrupts */
stat &= enable;
if (unlikely(!stat))
return IRQ_NONE;
if (stat & CA91CX42_LINT_DMA)
serviced |= ca91cx42_DMA_irqhandler(bridge);
if (stat & (CA91CX42_LINT_LM0 | CA91CX42_LINT_LM1 | CA91CX42_LINT_LM2 |
CA91CX42_LINT_LM3))
serviced |= ca91cx42_LM_irqhandler(bridge, stat);
if (stat & CA91CX42_LINT_MBOX)
serviced |= ca91cx42_MB_irqhandler(bridge, stat);
if (stat & CA91CX42_LINT_SW_IACK)
serviced |= ca91cx42_IACK_irqhandler(bridge);
if (stat & CA91CX42_LINT_VERR)
serviced |= ca91cx42_VERR_irqhandler(ca91cx42_bridge);
if (stat & CA91CX42_LINT_LERR)
serviced |= ca91cx42_LERR_irqhandler(ca91cx42_bridge);
if (stat & (CA91CX42_LINT_VIRQ1 | CA91CX42_LINT_VIRQ2 |
CA91CX42_LINT_VIRQ3 | CA91CX42_LINT_VIRQ4 |
CA91CX42_LINT_VIRQ5 | CA91CX42_LINT_VIRQ6 |
CA91CX42_LINT_VIRQ7))
serviced |= ca91cx42_VIRQ_irqhandler(ca91cx42_bridge, stat);
/* Clear serviced interrupts */
iowrite32(serviced, bridge->base + LINT_STAT);
return IRQ_HANDLED;
}
static int ca91cx42_irq_init(struct vme_bridge *ca91cx42_bridge)
{
int result, tmp;
struct pci_dev *pdev;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
/* Need pdev */
pdev = to_pci_dev(ca91cx42_bridge->parent);
/* Disable interrupts from PCI to VME */
iowrite32(0, bridge->base + VINT_EN);
/* Disable PCI interrupts */
iowrite32(0, bridge->base + LINT_EN);
/* Clear Any Pending PCI Interrupts */
iowrite32(0x00FFFFFF, bridge->base + LINT_STAT);
result = request_irq(pdev->irq, ca91cx42_irqhandler, IRQF_SHARED,
driver_name, ca91cx42_bridge);
if (result) {
dev_err(&pdev->dev, "Can't get assigned pci irq vector %02X\n",
pdev->irq);
return result;
}
/* Ensure all interrupts are mapped to PCI Interrupt 0 */
iowrite32(0, bridge->base + LINT_MAP0);
iowrite32(0, bridge->base + LINT_MAP1);
iowrite32(0, bridge->base + LINT_MAP2);
/* Enable DMA, mailbox & LM Interrupts */
tmp = CA91CX42_LINT_MBOX3 | CA91CX42_LINT_MBOX2 | CA91CX42_LINT_MBOX1 |
CA91CX42_LINT_MBOX0 | CA91CX42_LINT_SW_IACK |
CA91CX42_LINT_VERR | CA91CX42_LINT_LERR | CA91CX42_LINT_DMA;
iowrite32(tmp, bridge->base + LINT_EN);
return 0;
}
static void ca91cx42_irq_exit(struct ca91cx42_driver *bridge,
struct pci_dev *pdev)
{
struct vme_bridge *ca91cx42_bridge;
/* Disable interrupts from PCI to VME */
iowrite32(0, bridge->base + VINT_EN);
/* Disable PCI interrupts */
iowrite32(0, bridge->base + LINT_EN);
/* Clear Any Pending PCI Interrupts */
iowrite32(0x00FFFFFF, bridge->base + LINT_STAT);
ca91cx42_bridge = container_of((void *)bridge, struct vme_bridge,
driver_priv);
free_irq(pdev->irq, ca91cx42_bridge);
}
static int ca91cx42_iack_received(struct ca91cx42_driver *bridge, int level)
{
u32 tmp;
tmp = ioread32(bridge->base + LINT_STAT);
if (tmp & (1 << level))
return 0;
else
return 1;
}
/*
* Set up an VME interrupt
*/
static void ca91cx42_irq_set(struct vme_bridge *ca91cx42_bridge, int level,
int state, int sync)
{
struct pci_dev *pdev;
u32 tmp;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
/* Enable IRQ level */
tmp = ioread32(bridge->base + LINT_EN);
if (state == 0)
tmp &= ~CA91CX42_LINT_VIRQ[level];
else
tmp |= CA91CX42_LINT_VIRQ[level];
iowrite32(tmp, bridge->base + LINT_EN);
if ((state == 0) && (sync != 0)) {
pdev = to_pci_dev(ca91cx42_bridge->parent);
synchronize_irq(pdev->irq);
}
}
static int ca91cx42_irq_generate(struct vme_bridge *ca91cx42_bridge, int level,
int statid)
{
u32 tmp;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
/* Universe can only generate even vectors */
if (statid & 1)
return -EINVAL;
mutex_lock(&bridge->vme_int);
tmp = ioread32(bridge->base + VINT_EN);
/* Set Status/ID */
iowrite32(statid << 24, bridge->base + STATID);
/* Assert VMEbus IRQ */
tmp = tmp | (1 << (level + 24));
iowrite32(tmp, bridge->base + VINT_EN);
/* Wait for IACK */
wait_event_interruptible(bridge->iack_queue,
ca91cx42_iack_received(bridge, level));
/* Return interrupt to low state */
tmp = ioread32(bridge->base + VINT_EN);
tmp = tmp & ~(1 << (level + 24));
iowrite32(tmp, bridge->base + VINT_EN);
mutex_unlock(&bridge->vme_int);
return 0;
}
static int ca91cx42_slave_set(struct vme_slave_resource *image, int enabled,
unsigned long long vme_base, unsigned long long size,
dma_addr_t pci_base, u32 aspace, u32 cycle)
{
unsigned int i, addr = 0, granularity;
unsigned int temp_ctl = 0;
unsigned int vme_bound, pci_offset;
struct vme_bridge *ca91cx42_bridge;
struct ca91cx42_driver *bridge;
ca91cx42_bridge = image->parent;
bridge = ca91cx42_bridge->driver_priv;
i = image->number;
switch (aspace) {
case VME_A16:
addr |= CA91CX42_VSI_CTL_VAS_A16;
break;
case VME_A24:
addr |= CA91CX42_VSI_CTL_VAS_A24;
break;
case VME_A32:
addr |= CA91CX42_VSI_CTL_VAS_A32;
break;
case VME_USER1:
addr |= CA91CX42_VSI_CTL_VAS_USER1;
break;
case VME_USER2:
addr |= CA91CX42_VSI_CTL_VAS_USER2;
break;
case VME_A64:
case VME_CRCSR:
case VME_USER3:
case VME_USER4:
default:
dev_err(ca91cx42_bridge->parent, "Invalid address space\n");
return -EINVAL;
break;
}
/*
* Bound address is a valid address for the window, adjust
* accordingly
*/
vme_bound = vme_base + size;
pci_offset = pci_base - vme_base;
if ((i == 0) || (i == 4))
granularity = 0x1000;
else
granularity = 0x10000;
if (vme_base & (granularity - 1)) {
dev_err(ca91cx42_bridge->parent, "Invalid VME base "
"alignment\n");
return -EINVAL;
}
if (vme_bound & (granularity - 1)) {
dev_err(ca91cx42_bridge->parent, "Invalid VME bound "
"alignment\n");
return -EINVAL;
}
if (pci_offset & (granularity - 1)) {
dev_err(ca91cx42_bridge->parent, "Invalid PCI Offset "
"alignment\n");
return -EINVAL;
}
/* Disable while we are mucking around */
temp_ctl = ioread32(bridge->base + CA91CX42_VSI_CTL[i]);
temp_ctl &= ~CA91CX42_VSI_CTL_EN;
iowrite32(temp_ctl, bridge->base + CA91CX42_VSI_CTL[i]);
/* Setup mapping */
iowrite32(vme_base, bridge->base + CA91CX42_VSI_BS[i]);
iowrite32(vme_bound, bridge->base + CA91CX42_VSI_BD[i]);
iowrite32(pci_offset, bridge->base + CA91CX42_VSI_TO[i]);
/* Setup address space */
temp_ctl &= ~CA91CX42_VSI_CTL_VAS_M;
temp_ctl |= addr;
/* Setup cycle types */
temp_ctl &= ~(CA91CX42_VSI_CTL_PGM_M | CA91CX42_VSI_CTL_SUPER_M);
if (cycle & VME_SUPER)
temp_ctl |= CA91CX42_VSI_CTL_SUPER_SUPR;
if (cycle & VME_USER)
temp_ctl |= CA91CX42_VSI_CTL_SUPER_NPRIV;
if (cycle & VME_PROG)
temp_ctl |= CA91CX42_VSI_CTL_PGM_PGM;
if (cycle & VME_DATA)
temp_ctl |= CA91CX42_VSI_CTL_PGM_DATA;
/* Write ctl reg without enable */
iowrite32(temp_ctl, bridge->base + CA91CX42_VSI_CTL[i]);
if (enabled)
temp_ctl |= CA91CX42_VSI_CTL_EN;
iowrite32(temp_ctl, bridge->base + CA91CX42_VSI_CTL[i]);
return 0;
}
static int ca91cx42_slave_get(struct vme_slave_resource *image, int *enabled,
unsigned long long *vme_base, unsigned long long *size,
dma_addr_t *pci_base, u32 *aspace, u32 *cycle)
{
unsigned int i, granularity = 0, ctl = 0;
unsigned long long vme_bound, pci_offset;
struct ca91cx42_driver *bridge;
bridge = image->parent->driver_priv;
i = image->number;
if ((i == 0) || (i == 4))
granularity = 0x1000;
else
granularity = 0x10000;
/* Read Registers */
ctl = ioread32(bridge->base + CA91CX42_VSI_CTL[i]);
*vme_base = ioread32(bridge->base + CA91CX42_VSI_BS[i]);
vme_bound = ioread32(bridge->base + CA91CX42_VSI_BD[i]);
pci_offset = ioread32(bridge->base + CA91CX42_VSI_TO[i]);
*pci_base = (dma_addr_t)*vme_base + pci_offset;
*size = (unsigned long long)((vme_bound - *vme_base) + granularity);
*enabled = 0;
*aspace = 0;
*cycle = 0;
if (ctl & CA91CX42_VSI_CTL_EN)
*enabled = 1;
if ((ctl & CA91CX42_VSI_CTL_VAS_M) == CA91CX42_VSI_CTL_VAS_A16)
*aspace = VME_A16;
if ((ctl & CA91CX42_VSI_CTL_VAS_M) == CA91CX42_VSI_CTL_VAS_A24)
*aspace = VME_A24;
if ((ctl & CA91CX42_VSI_CTL_VAS_M) == CA91CX42_VSI_CTL_VAS_A32)
*aspace = VME_A32;
if ((ctl & CA91CX42_VSI_CTL_VAS_M) == CA91CX42_VSI_CTL_VAS_USER1)
*aspace = VME_USER1;
if ((ctl & CA91CX42_VSI_CTL_VAS_M) == CA91CX42_VSI_CTL_VAS_USER2)
*aspace = VME_USER2;
if (ctl & CA91CX42_VSI_CTL_SUPER_SUPR)
*cycle |= VME_SUPER;
if (ctl & CA91CX42_VSI_CTL_SUPER_NPRIV)
*cycle |= VME_USER;
if (ctl & CA91CX42_VSI_CTL_PGM_PGM)
*cycle |= VME_PROG;
if (ctl & CA91CX42_VSI_CTL_PGM_DATA)
*cycle |= VME_DATA;
return 0;
}
/*
* Allocate and map PCI Resource
*/
static int ca91cx42_alloc_resource(struct vme_master_resource *image,
unsigned long long size)
{
unsigned long long existing_size;
int retval = 0;
struct pci_dev *pdev;
struct vme_bridge *ca91cx42_bridge;
ca91cx42_bridge = image->parent;
/* Find pci_dev container of dev */
if (!ca91cx42_bridge->parent) {
dev_err(ca91cx42_bridge->parent, "Dev entry NULL\n");
return -EINVAL;
}
pdev = to_pci_dev(ca91cx42_bridge->parent);
existing_size = (unsigned long long)(image->bus_resource.end -
image->bus_resource.start);
/* If the existing size is OK, return */
if (existing_size == (size - 1))
return 0;
if (existing_size != 0) {
iounmap(image->kern_base);
image->kern_base = NULL;
kfree(image->bus_resource.name);
release_resource(&image->bus_resource);
memset(&image->bus_resource, 0, sizeof(image->bus_resource));
}
if (!image->bus_resource.name) {
image->bus_resource.name = kmalloc(VMENAMSIZ+3, GFP_ATOMIC);
if (!image->bus_resource.name) {
retval = -ENOMEM;
goto err_name;
}
}
sprintf((char *)image->bus_resource.name, "%s.%d",
ca91cx42_bridge->name, image->number);
image->bus_resource.start = 0;
image->bus_resource.end = (unsigned long)size;
image->bus_resource.flags = IORESOURCE_MEM;
retval = pci_bus_alloc_resource(pdev->bus,
&image->bus_resource, size, 0x10000, PCIBIOS_MIN_MEM,
0, NULL, NULL);
if (retval) {
dev_err(ca91cx42_bridge->parent, "Failed to allocate mem "
"resource for window %d size 0x%lx start 0x%lx\n",
image->number, (unsigned long)size,
(unsigned long)image->bus_resource.start);
goto err_resource;
}
image->kern_base = ioremap_nocache(
image->bus_resource.start, size);
if (!image->kern_base) {
dev_err(ca91cx42_bridge->parent, "Failed to remap resource\n");
retval = -ENOMEM;
goto err_remap;
}
return 0;
err_remap:
release_resource(&image->bus_resource);
err_resource:
kfree(image->bus_resource.name);
memset(&image->bus_resource, 0, sizeof(image->bus_resource));
err_name:
return retval;
}
/*
* Free and unmap PCI Resource
*/
static void ca91cx42_free_resource(struct vme_master_resource *image)
{
iounmap(image->kern_base);
image->kern_base = NULL;
release_resource(&image->bus_resource);
kfree(image->bus_resource.name);
memset(&image->bus_resource, 0, sizeof(image->bus_resource));
}
static int ca91cx42_master_set(struct vme_master_resource *image, int enabled,
unsigned long long vme_base, unsigned long long size, u32 aspace,
u32 cycle, u32 dwidth)
{
int retval = 0;
unsigned int i, granularity = 0;
unsigned int temp_ctl = 0;
unsigned long long pci_bound, vme_offset, pci_base;
struct vme_bridge *ca91cx42_bridge;
struct ca91cx42_driver *bridge;
ca91cx42_bridge = image->parent;
bridge = ca91cx42_bridge->driver_priv;
i = image->number;
if ((i == 0) || (i == 4))
granularity = 0x1000;
else
granularity = 0x10000;
/* Verify input data */
if (vme_base & (granularity - 1)) {
dev_err(ca91cx42_bridge->parent, "Invalid VME Window "
"alignment\n");
retval = -EINVAL;
goto err_window;
}
if (size & (granularity - 1)) {
dev_err(ca91cx42_bridge->parent, "Invalid VME Window "
"alignment\n");
retval = -EINVAL;
goto err_window;
}
spin_lock(&image->lock);
/*
* Let's allocate the resource here rather than further up the stack as
* it avoids pushing loads of bus dependent stuff up the stack
*/
retval = ca91cx42_alloc_resource(image, size);
if (retval) {
spin_unlock(&image->lock);
dev_err(ca91cx42_bridge->parent, "Unable to allocate memory "
"for resource name\n");
retval = -ENOMEM;
goto err_res;
}
pci_base = (unsigned long long)image->bus_resource.start;
/*
* Bound address is a valid address for the window, adjust
* according to window granularity.
*/
pci_bound = pci_base + size;
vme_offset = vme_base - pci_base;
/* Disable while we are mucking around */
temp_ctl = ioread32(bridge->base + CA91CX42_LSI_CTL[i]);
temp_ctl &= ~CA91CX42_LSI_CTL_EN;
iowrite32(temp_ctl, bridge->base + CA91CX42_LSI_CTL[i]);
/* Setup cycle types */
temp_ctl &= ~CA91CX42_LSI_CTL_VCT_M;
if (cycle & VME_BLT)
temp_ctl |= CA91CX42_LSI_CTL_VCT_BLT;
if (cycle & VME_MBLT)
temp_ctl |= CA91CX42_LSI_CTL_VCT_MBLT;
/* Setup data width */
temp_ctl &= ~CA91CX42_LSI_CTL_VDW_M;
switch (dwidth) {
case VME_D8:
temp_ctl |= CA91CX42_LSI_CTL_VDW_D8;
break;
case VME_D16:
temp_ctl |= CA91CX42_LSI_CTL_VDW_D16;
break;
case VME_D32:
temp_ctl |= CA91CX42_LSI_CTL_VDW_D32;
break;
case VME_D64:
temp_ctl |= CA91CX42_LSI_CTL_VDW_D64;
break;
default:
spin_unlock(&image->lock);
dev_err(ca91cx42_bridge->parent, "Invalid data width\n");
retval = -EINVAL;
goto err_dwidth;
break;
}
/* Setup address space */
temp_ctl &= ~CA91CX42_LSI_CTL_VAS_M;
switch (aspace) {
case VME_A16:
temp_ctl |= CA91CX42_LSI_CTL_VAS_A16;
break;
case VME_A24:
temp_ctl |= CA91CX42_LSI_CTL_VAS_A24;
break;
case VME_A32:
temp_ctl |= CA91CX42_LSI_CTL_VAS_A32;
break;
case VME_CRCSR:
temp_ctl |= CA91CX42_LSI_CTL_VAS_CRCSR;
break;
case VME_USER1:
temp_ctl |= CA91CX42_LSI_CTL_VAS_USER1;
break;
case VME_USER2:
temp_ctl |= CA91CX42_LSI_CTL_VAS_USER2;
break;
case VME_A64:
case VME_USER3:
case VME_USER4:
default:
spin_unlock(&image->lock);
dev_err(ca91cx42_bridge->parent, "Invalid address space\n");
retval = -EINVAL;
goto err_aspace;
break;
}
temp_ctl &= ~(CA91CX42_LSI_CTL_PGM_M | CA91CX42_LSI_CTL_SUPER_M);
if (cycle & VME_SUPER)
temp_ctl |= CA91CX42_LSI_CTL_SUPER_SUPR;
if (cycle & VME_PROG)
temp_ctl |= CA91CX42_LSI_CTL_PGM_PGM;
/* Setup mapping */
iowrite32(pci_base, bridge->base + CA91CX42_LSI_BS[i]);
iowrite32(pci_bound, bridge->base + CA91CX42_LSI_BD[i]);
iowrite32(vme_offset, bridge->base + CA91CX42_LSI_TO[i]);
/* Write ctl reg without enable */
iowrite32(temp_ctl, bridge->base + CA91CX42_LSI_CTL[i]);
if (enabled)
temp_ctl |= CA91CX42_LSI_CTL_EN;
iowrite32(temp_ctl, bridge->base + CA91CX42_LSI_CTL[i]);
spin_unlock(&image->lock);
return 0;
err_aspace:
err_dwidth:
ca91cx42_free_resource(image);
err_res:
err_window:
return retval;
}
static int __ca91cx42_master_get(struct vme_master_resource *image,
int *enabled, unsigned long long *vme_base, unsigned long long *size,
u32 *aspace, u32 *cycle, u32 *dwidth)
{
unsigned int i, ctl;
unsigned long long pci_base, pci_bound, vme_offset;
struct ca91cx42_driver *bridge;
bridge = image->parent->driver_priv;
i = image->number;
ctl = ioread32(bridge->base + CA91CX42_LSI_CTL[i]);
pci_base = ioread32(bridge->base + CA91CX42_LSI_BS[i]);
vme_offset = ioread32(bridge->base + CA91CX42_LSI_TO[i]);
pci_bound = ioread32(bridge->base + CA91CX42_LSI_BD[i]);
*vme_base = pci_base + vme_offset;
*size = (unsigned long long)(pci_bound - pci_base);
*enabled = 0;
*aspace = 0;
*cycle = 0;
*dwidth = 0;
if (ctl & CA91CX42_LSI_CTL_EN)
*enabled = 1;
/* Setup address space */
switch (ctl & CA91CX42_LSI_CTL_VAS_M) {
case CA91CX42_LSI_CTL_VAS_A16:
*aspace = VME_A16;
break;
case CA91CX42_LSI_CTL_VAS_A24:
*aspace = VME_A24;
break;
case CA91CX42_LSI_CTL_VAS_A32:
*aspace = VME_A32;
break;
case CA91CX42_LSI_CTL_VAS_CRCSR:
*aspace = VME_CRCSR;
break;
case CA91CX42_LSI_CTL_VAS_USER1:
*aspace = VME_USER1;
break;
case CA91CX42_LSI_CTL_VAS_USER2:
*aspace = VME_USER2;
break;
}
/* XXX Not sure howto check for MBLT */
/* Setup cycle types */
if (ctl & CA91CX42_LSI_CTL_VCT_BLT)
*cycle |= VME_BLT;
else
*cycle |= VME_SCT;
if (ctl & CA91CX42_LSI_CTL_SUPER_SUPR)
*cycle |= VME_SUPER;
else
*cycle |= VME_USER;
if (ctl & CA91CX42_LSI_CTL_PGM_PGM)
*cycle = VME_PROG;
else
*cycle = VME_DATA;
/* Setup data width */
switch (ctl & CA91CX42_LSI_CTL_VDW_M) {
case CA91CX42_LSI_CTL_VDW_D8:
*dwidth = VME_D8;
break;
case CA91CX42_LSI_CTL_VDW_D16:
*dwidth = VME_D16;
break;
case CA91CX42_LSI_CTL_VDW_D32:
*dwidth = VME_D32;
break;
case CA91CX42_LSI_CTL_VDW_D64:
*dwidth = VME_D64;
break;
}
return 0;
}
static int ca91cx42_master_get(struct vme_master_resource *image, int *enabled,
unsigned long long *vme_base, unsigned long long *size, u32 *aspace,
u32 *cycle, u32 *dwidth)
{
int retval;
spin_lock(&image->lock);
retval = __ca91cx42_master_get(image, enabled, vme_base, size, aspace,
cycle, dwidth);
spin_unlock(&image->lock);
return retval;
}
static ssize_t ca91cx42_master_read(struct vme_master_resource *image,
void *buf, size_t count, loff_t offset)
{
ssize_t retval;
void __iomem *addr = image->kern_base + offset;
unsigned int done = 0;
unsigned int count32;
if (count == 0)
return 0;
spin_lock(&image->lock);
/* The following code handles VME address alignment. We cannot use
* memcpy_xxx here because it may cut data transfers in to 8-bit
* cycles when D16 or D32 cycles are required on the VME bus.
* On the other hand, the bridge itself assures that the maximum data
* cycle configured for the transfer is used and splits it
* automatically for non-aligned addresses, so we don't want the
* overhead of needlessly forcing small transfers for the entire cycle.
*/
if ((uintptr_t)addr & 0x1) {
*(u8 *)buf = ioread8(addr);
done += 1;
if (done == count)
goto out;
}
if ((uintptr_t)(addr + done) & 0x2) {
if ((count - done) < 2) {
*(u8 *)(buf + done) = ioread8(addr + done);
done += 1;
goto out;
} else {
*(u16 *)(buf + done) = ioread16(addr + done);
done += 2;
}
}
count32 = (count - done) & ~0x3;
while (done < count32) {
*(u32 *)(buf + done) = ioread32(addr + done);
done += 4;
}
if ((count - done) & 0x2) {
*(u16 *)(buf + done) = ioread16(addr + done);
done += 2;
}
if ((count - done) & 0x1) {
*(u8 *)(buf + done) = ioread8(addr + done);
done += 1;
}
out:
retval = count;
spin_unlock(&image->lock);
return retval;
}
static ssize_t ca91cx42_master_write(struct vme_master_resource *image,
void *buf, size_t count, loff_t offset)
{
ssize_t retval;
void __iomem *addr = image->kern_base + offset;
unsigned int done = 0;
unsigned int count32;
if (count == 0)
return 0;
spin_lock(&image->lock);
/* Here we apply for the same strategy we do in master_read
* function in order to assure the correct cycles.
*/
if ((uintptr_t)addr & 0x1) {
iowrite8(*(u8 *)buf, addr);
done += 1;
if (done == count)
goto out;
}
if ((uintptr_t)(addr + done) & 0x2) {
if ((count - done) < 2) {
iowrite8(*(u8 *)(buf + done), addr + done);
done += 1;
goto out;
} else {
iowrite16(*(u16 *)(buf + done), addr + done);
done += 2;
}
}
count32 = (count - done) & ~0x3;
while (done < count32) {
iowrite32(*(u32 *)(buf + done), addr + done);
done += 4;
}
if ((count - done) & 0x2) {
iowrite16(*(u16 *)(buf + done), addr + done);
done += 2;
}
if ((count - done) & 0x1) {
iowrite8(*(u8 *)(buf + done), addr + done);
done += 1;
}
out:
retval = count;
spin_unlock(&image->lock);
return retval;
}
static unsigned int ca91cx42_master_rmw(struct vme_master_resource *image,
unsigned int mask, unsigned int compare, unsigned int swap,
loff_t offset)
{
u32 result;
uintptr_t pci_addr;
int i;
struct ca91cx42_driver *bridge;
struct device *dev;
bridge = image->parent->driver_priv;
dev = image->parent->parent;
/* Find the PCI address that maps to the desired VME address */
i = image->number;
/* Locking as we can only do one of these at a time */
mutex_lock(&bridge->vme_rmw);
/* Lock image */
spin_lock(&image->lock);
pci_addr = (uintptr_t)image->kern_base + offset;
/* Address must be 4-byte aligned */
if (pci_addr & 0x3) {
dev_err(dev, "RMW Address not 4-byte aligned\n");
result = -EINVAL;
goto out;
}
/* Ensure RMW Disabled whilst configuring */
iowrite32(0, bridge->base + SCYC_CTL);
/* Configure registers */
iowrite32(mask, bridge->base + SCYC_EN);
iowrite32(compare, bridge->base + SCYC_CMP);
iowrite32(swap, bridge->base + SCYC_SWP);
iowrite32(pci_addr, bridge->base + SCYC_ADDR);
/* Enable RMW */
iowrite32(CA91CX42_SCYC_CTL_CYC_RMW, bridge->base + SCYC_CTL);
/* Kick process off with a read to the required address. */
result = ioread32(image->kern_base + offset);
/* Disable RMW */
iowrite32(0, bridge->base + SCYC_CTL);
out:
spin_unlock(&image->lock);
mutex_unlock(&bridge->vme_rmw);
return result;
}
static int ca91cx42_dma_list_add(struct vme_dma_list *list,
struct vme_dma_attr *src, struct vme_dma_attr *dest, size_t count)
{
struct ca91cx42_dma_entry *entry, *prev;
struct vme_dma_pci *pci_attr;
struct vme_dma_vme *vme_attr;
dma_addr_t desc_ptr;
int retval = 0;
struct device *dev;
dev = list->parent->parent->parent;
/* XXX descriptor must be aligned on 64-bit boundaries */
entry = kmalloc(sizeof(*entry), GFP_KERNEL);
if (!entry) {
retval = -ENOMEM;
goto err_mem;
}
/* Test descriptor alignment */
if ((unsigned long)&entry->descriptor & CA91CX42_DCPP_M) {
dev_err(dev, "Descriptor not aligned to 16 byte boundary as "
"required: %p\n", &entry->descriptor);
retval = -EINVAL;
goto err_align;
}
memset(&entry->descriptor, 0, sizeof(entry->descriptor));
if (dest->type == VME_DMA_VME) {
entry->descriptor.dctl |= CA91CX42_DCTL_L2V;
vme_attr = dest->private;
pci_attr = src->private;
} else {
vme_attr = src->private;
pci_attr = dest->private;
}
/* Check we can do fulfill required attributes */
if ((vme_attr->aspace & ~(VME_A16 | VME_A24 | VME_A32 | VME_USER1 |
VME_USER2)) != 0) {
dev_err(dev, "Unsupported cycle type\n");
retval = -EINVAL;
goto err_aspace;
}
if ((vme_attr->cycle & ~(VME_SCT | VME_BLT | VME_SUPER | VME_USER |
VME_PROG | VME_DATA)) != 0) {
dev_err(dev, "Unsupported cycle type\n");
retval = -EINVAL;
goto err_cycle;
}
/* Check to see if we can fulfill source and destination */
if (!(((src->type == VME_DMA_PCI) && (dest->type == VME_DMA_VME)) ||
((src->type == VME_DMA_VME) && (dest->type == VME_DMA_PCI)))) {
dev_err(dev, "Cannot perform transfer with this "
"source-destination combination\n");
retval = -EINVAL;
goto err_direct;
}
/* Setup cycle types */
if (vme_attr->cycle & VME_BLT)
entry->descriptor.dctl |= CA91CX42_DCTL_VCT_BLT;
/* Setup data width */
switch (vme_attr->dwidth) {
case VME_D8:
entry->descriptor.dctl |= CA91CX42_DCTL_VDW_D8;
break;
case VME_D16:
entry->descriptor.dctl |= CA91CX42_DCTL_VDW_D16;
break;
case VME_D32:
entry->descriptor.dctl |= CA91CX42_DCTL_VDW_D32;
break;
case VME_D64:
entry->descriptor.dctl |= CA91CX42_DCTL_VDW_D64;
break;
default:
dev_err(dev, "Invalid data width\n");
return -EINVAL;
}
/* Setup address space */
switch (vme_attr->aspace) {
case VME_A16:
entry->descriptor.dctl |= CA91CX42_DCTL_VAS_A16;
break;
case VME_A24:
entry->descriptor.dctl |= CA91CX42_DCTL_VAS_A24;
break;
case VME_A32:
entry->descriptor.dctl |= CA91CX42_DCTL_VAS_A32;
break;
case VME_USER1:
entry->descriptor.dctl |= CA91CX42_DCTL_VAS_USER1;
break;
case VME_USER2:
entry->descriptor.dctl |= CA91CX42_DCTL_VAS_USER2;
break;
default:
dev_err(dev, "Invalid address space\n");
return -EINVAL;
break;
}
if (vme_attr->cycle & VME_SUPER)
entry->descriptor.dctl |= CA91CX42_DCTL_SUPER_SUPR;
if (vme_attr->cycle & VME_PROG)
entry->descriptor.dctl |= CA91CX42_DCTL_PGM_PGM;
entry->descriptor.dtbc = count;
entry->descriptor.dla = pci_attr->address;
entry->descriptor.dva = vme_attr->address;
entry->descriptor.dcpp = CA91CX42_DCPP_NULL;
/* Add to list */
list_add_tail(&entry->list, &list->entries);
/* Fill out previous descriptors "Next Address" */
if (entry->list.prev != &list->entries) {
prev = list_entry(entry->list.prev, struct ca91cx42_dma_entry,
list);
/* We need the bus address for the pointer */
desc_ptr = virt_to_bus(&entry->descriptor);
prev->descriptor.dcpp = desc_ptr & ~CA91CX42_DCPP_M;
}
return 0;
err_cycle:
err_aspace:
err_direct:
err_align:
kfree(entry);
err_mem:
return retval;
}
static int ca91cx42_dma_busy(struct vme_bridge *ca91cx42_bridge)
{
u32 tmp;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
tmp = ioread32(bridge->base + DGCS);
if (tmp & CA91CX42_DGCS_ACT)
return 0;
else
return 1;
}
static int ca91cx42_dma_list_exec(struct vme_dma_list *list)
{
struct vme_dma_resource *ctrlr;
struct ca91cx42_dma_entry *entry;
int retval;
dma_addr_t bus_addr;
u32 val;
struct device *dev;
struct ca91cx42_driver *bridge;
ctrlr = list->parent;
bridge = ctrlr->parent->driver_priv;
dev = ctrlr->parent->parent;
mutex_lock(&ctrlr->mtx);
if (!(list_empty(&ctrlr->running))) {
/*
* XXX We have an active DMA transfer and currently haven't
* sorted out the mechanism for "pending" DMA transfers.
* Return busy.
*/
/* Need to add to pending here */
mutex_unlock(&ctrlr->mtx);
return -EBUSY;
} else {
list_add(&list->list, &ctrlr->running);
}
/* Get first bus address and write into registers */
entry = list_first_entry(&list->entries, struct ca91cx42_dma_entry,
list);
bus_addr = virt_to_bus(&entry->descriptor);
mutex_unlock(&ctrlr->mtx);
iowrite32(0, bridge->base + DTBC);
iowrite32(bus_addr & ~CA91CX42_DCPP_M, bridge->base + DCPP);
/* Start the operation */
val = ioread32(bridge->base + DGCS);
/* XXX Could set VMEbus On and Off Counters here */
val &= (CA91CX42_DGCS_VON_M | CA91CX42_DGCS_VOFF_M);
val |= (CA91CX42_DGCS_CHAIN | CA91CX42_DGCS_STOP | CA91CX42_DGCS_HALT |
CA91CX42_DGCS_DONE | CA91CX42_DGCS_LERR | CA91CX42_DGCS_VERR |
CA91CX42_DGCS_PERR);
iowrite32(val, bridge->base + DGCS);
val |= CA91CX42_DGCS_GO;
iowrite32(val, bridge->base + DGCS);
retval = wait_event_interruptible(bridge->dma_queue,
ca91cx42_dma_busy(ctrlr->parent));
if (retval) {
val = ioread32(bridge->base + DGCS);
iowrite32(val | CA91CX42_DGCS_STOP_REQ, bridge->base + DGCS);
/* Wait for the operation to abort */
wait_event(bridge->dma_queue,
ca91cx42_dma_busy(ctrlr->parent));
retval = -EINTR;
goto exit;
}
/*
* Read status register, this register is valid until we kick off a
* new transfer.
*/
val = ioread32(bridge->base + DGCS);
if (val & (CA91CX42_DGCS_LERR | CA91CX42_DGCS_VERR |
CA91CX42_DGCS_PERR)) {
dev_err(dev, "ca91c042: DMA Error. DGCS=%08X\n", val);
val = ioread32(bridge->base + DCTL);
retval = -EIO;
}
exit:
/* Remove list from running list */
mutex_lock(&ctrlr->mtx);
list_del(&list->list);
mutex_unlock(&ctrlr->mtx);
return retval;
}
static int ca91cx42_dma_list_empty(struct vme_dma_list *list)
{
struct list_head *pos, *temp;
struct ca91cx42_dma_entry *entry;
/* detach and free each entry */
list_for_each_safe(pos, temp, &list->entries) {
list_del(pos);
entry = list_entry(pos, struct ca91cx42_dma_entry, list);
kfree(entry);
}
return 0;
}
/*
* All 4 location monitors reside at the same base - this is therefore a
* system wide configuration.
*
* This does not enable the LM monitor - that should be done when the first
* callback is attached and disabled when the last callback is removed.
*/
static int ca91cx42_lm_set(struct vme_lm_resource *lm,
unsigned long long lm_base, u32 aspace, u32 cycle)
{
u32 temp_base, lm_ctl = 0;
int i;
struct ca91cx42_driver *bridge;
struct device *dev;
bridge = lm->parent->driver_priv;
dev = lm->parent->parent;
/* Check the alignment of the location monitor */
temp_base = (u32)lm_base;
if (temp_base & 0xffff) {
dev_err(dev, "Location monitor must be aligned to 64KB "
"boundary");
return -EINVAL;
}
mutex_lock(&lm->mtx);
/* If we already have a callback attached, we can't move it! */
for (i = 0; i < lm->monitors; i++) {
if (bridge->lm_callback[i]) {
mutex_unlock(&lm->mtx);
dev_err(dev, "Location monitor callback attached, "
"can't reset\n");
return -EBUSY;
}
}
switch (aspace) {
case VME_A16:
lm_ctl |= CA91CX42_LM_CTL_AS_A16;
break;
case VME_A24:
lm_ctl |= CA91CX42_LM_CTL_AS_A24;
break;
case VME_A32:
lm_ctl |= CA91CX42_LM_CTL_AS_A32;
break;
default:
mutex_unlock(&lm->mtx);
dev_err(dev, "Invalid address space\n");
return -EINVAL;
break;
}
if (cycle & VME_SUPER)
lm_ctl |= CA91CX42_LM_CTL_SUPR;
if (cycle & VME_USER)
lm_ctl |= CA91CX42_LM_CTL_NPRIV;
if (cycle & VME_PROG)
lm_ctl |= CA91CX42_LM_CTL_PGM;
if (cycle & VME_DATA)
lm_ctl |= CA91CX42_LM_CTL_DATA;
iowrite32(lm_base, bridge->base + LM_BS);
iowrite32(lm_ctl, bridge->base + LM_CTL);
mutex_unlock(&lm->mtx);
return 0;
}
/* Get configuration of the callback monitor and return whether it is enabled
* or disabled.
*/
static int ca91cx42_lm_get(struct vme_lm_resource *lm,
unsigned long long *lm_base, u32 *aspace, u32 *cycle)
{
u32 lm_ctl, enabled = 0;
struct ca91cx42_driver *bridge;
bridge = lm->parent->driver_priv;
mutex_lock(&lm->mtx);
*lm_base = (unsigned long long)ioread32(bridge->base + LM_BS);
lm_ctl = ioread32(bridge->base + LM_CTL);
if (lm_ctl & CA91CX42_LM_CTL_EN)
enabled = 1;
if ((lm_ctl & CA91CX42_LM_CTL_AS_M) == CA91CX42_LM_CTL_AS_A16)
*aspace = VME_A16;
if ((lm_ctl & CA91CX42_LM_CTL_AS_M) == CA91CX42_LM_CTL_AS_A24)
*aspace = VME_A24;
if ((lm_ctl & CA91CX42_LM_CTL_AS_M) == CA91CX42_LM_CTL_AS_A32)
*aspace = VME_A32;
*cycle = 0;
if (lm_ctl & CA91CX42_LM_CTL_SUPR)
*cycle |= VME_SUPER;
if (lm_ctl & CA91CX42_LM_CTL_NPRIV)
*cycle |= VME_USER;
if (lm_ctl & CA91CX42_LM_CTL_PGM)
*cycle |= VME_PROG;
if (lm_ctl & CA91CX42_LM_CTL_DATA)
*cycle |= VME_DATA;
mutex_unlock(&lm->mtx);
return enabled;
}
/*
* Attach a callback to a specific location monitor.
*
* Callback will be passed the monitor triggered.
*/
static int ca91cx42_lm_attach(struct vme_lm_resource *lm, int monitor,
void (*callback)(void *), void *data)
{
u32 lm_ctl, tmp;
struct ca91cx42_driver *bridge;
struct device *dev;
bridge = lm->parent->driver_priv;
dev = lm->parent->parent;
mutex_lock(&lm->mtx);
/* Ensure that the location monitor is configured - need PGM or DATA */
lm_ctl = ioread32(bridge->base + LM_CTL);
if ((lm_ctl & (CA91CX42_LM_CTL_PGM | CA91CX42_LM_CTL_DATA)) == 0) {
mutex_unlock(&lm->mtx);
dev_err(dev, "Location monitor not properly configured\n");
return -EINVAL;
}
/* Check that a callback isn't already attached */
if (bridge->lm_callback[monitor]) {
mutex_unlock(&lm->mtx);
dev_err(dev, "Existing callback attached\n");
return -EBUSY;
}
/* Attach callback */
bridge->lm_callback[monitor] = callback;
bridge->lm_data[monitor] = data;
/* Enable Location Monitor interrupt */
tmp = ioread32(bridge->base + LINT_EN);
tmp |= CA91CX42_LINT_LM[monitor];
iowrite32(tmp, bridge->base + LINT_EN);
/* Ensure that global Location Monitor Enable set */
if ((lm_ctl & CA91CX42_LM_CTL_EN) == 0) {
lm_ctl |= CA91CX42_LM_CTL_EN;
iowrite32(lm_ctl, bridge->base + LM_CTL);
}
mutex_unlock(&lm->mtx);
return 0;
}
/*
* Detach a callback function forn a specific location monitor.
*/
static int ca91cx42_lm_detach(struct vme_lm_resource *lm, int monitor)
{
u32 tmp;
struct ca91cx42_driver *bridge;
bridge = lm->parent->driver_priv;
mutex_lock(&lm->mtx);
/* Disable Location Monitor and ensure previous interrupts are clear */
tmp = ioread32(bridge->base + LINT_EN);
tmp &= ~CA91CX42_LINT_LM[monitor];
iowrite32(tmp, bridge->base + LINT_EN);
iowrite32(CA91CX42_LINT_LM[monitor],
bridge->base + LINT_STAT);
/* Detach callback */
bridge->lm_callback[monitor] = NULL;
bridge->lm_data[monitor] = NULL;
/* If all location monitors disabled, disable global Location Monitor */
if ((tmp & (CA91CX42_LINT_LM0 | CA91CX42_LINT_LM1 | CA91CX42_LINT_LM2 |
CA91CX42_LINT_LM3)) == 0) {
tmp = ioread32(bridge->base + LM_CTL);
tmp &= ~CA91CX42_LM_CTL_EN;
iowrite32(tmp, bridge->base + LM_CTL);
}
mutex_unlock(&lm->mtx);
return 0;
}
static int ca91cx42_slot_get(struct vme_bridge *ca91cx42_bridge)
{
u32 slot = 0;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
if (!geoid) {
slot = ioread32(bridge->base + VCSR_BS);
slot = ((slot & CA91CX42_VCSR_BS_SLOT_M) >> 27);
} else
slot = geoid;
return (int)slot;
}
static void *ca91cx42_alloc_consistent(struct device *parent, size_t size,
dma_addr_t *dma)
{
struct pci_dev *pdev;
/* Find pci_dev container of dev */
pdev = to_pci_dev(parent);
return pci_alloc_consistent(pdev, size, dma);
}
static void ca91cx42_free_consistent(struct device *parent, size_t size,
void *vaddr, dma_addr_t dma)
{
struct pci_dev *pdev;
/* Find pci_dev container of dev */
pdev = to_pci_dev(parent);
pci_free_consistent(pdev, size, vaddr, dma);
}
/*
* Configure CR/CSR space
*
* Access to the CR/CSR can be configured at power-up. The location of the
* CR/CSR registers in the CR/CSR address space is determined by the boards
* Auto-ID or Geographic address. This function ensures that the window is
* enabled at an offset consistent with the boards geopgraphic address.
*/
static int ca91cx42_crcsr_init(struct vme_bridge *ca91cx42_bridge,
struct pci_dev *pdev)
{
unsigned int crcsr_addr;
int tmp, slot;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
slot = ca91cx42_slot_get(ca91cx42_bridge);
/* Write CSR Base Address if slot ID is supplied as a module param */
if (geoid)
iowrite32(geoid << 27, bridge->base + VCSR_BS);
dev_info(&pdev->dev, "CR/CSR Offset: %d\n", slot);
if (slot == 0) {
dev_err(&pdev->dev, "Slot number is unset, not configuring "
"CR/CSR space\n");
return -EINVAL;
}
/* Allocate mem for CR/CSR image */
bridge->crcsr_kernel = pci_zalloc_consistent(pdev, VME_CRCSR_BUF_SIZE,
&bridge->crcsr_bus);
if (!bridge->crcsr_kernel) {
dev_err(&pdev->dev, "Failed to allocate memory for CR/CSR "
"image\n");
return -ENOMEM;
}
crcsr_addr = slot * (512 * 1024);
iowrite32(bridge->crcsr_bus - crcsr_addr, bridge->base + VCSR_TO);
tmp = ioread32(bridge->base + VCSR_CTL);
tmp |= CA91CX42_VCSR_CTL_EN;
iowrite32(tmp, bridge->base + VCSR_CTL);
return 0;
}
static void ca91cx42_crcsr_exit(struct vme_bridge *ca91cx42_bridge,
struct pci_dev *pdev)
{
u32 tmp;
struct ca91cx42_driver *bridge;
bridge = ca91cx42_bridge->driver_priv;
/* Turn off CR/CSR space */
tmp = ioread32(bridge->base + VCSR_CTL);
tmp &= ~CA91CX42_VCSR_CTL_EN;
iowrite32(tmp, bridge->base + VCSR_CTL);
/* Free image */
iowrite32(0, bridge->base + VCSR_TO);
pci_free_consistent(pdev, VME_CRCSR_BUF_SIZE, bridge->crcsr_kernel,
bridge->crcsr_bus);
}
static int ca91cx42_probe(struct pci_dev *pdev, const struct pci_device_id *id)
{
int retval, i;
u32 data;
struct list_head *pos = NULL, *n;
struct vme_bridge *ca91cx42_bridge;
struct ca91cx42_driver *ca91cx42_device;
struct vme_master_resource *master_image;
struct vme_slave_resource *slave_image;
struct vme_dma_resource *dma_ctrlr;
struct vme_lm_resource *lm;
/* We want to support more than one of each bridge so we need to
* dynamically allocate the bridge structure
*/
ca91cx42_bridge = kzalloc(sizeof(*ca91cx42_bridge), GFP_KERNEL);
if (!ca91cx42_bridge) {
retval = -ENOMEM;
goto err_struct;
}
vme_init_bridge(ca91cx42_bridge);
ca91cx42_device = kzalloc(sizeof(*ca91cx42_device), GFP_KERNEL);
if (!ca91cx42_device) {
retval = -ENOMEM;
goto err_driver;
}
ca91cx42_bridge->driver_priv = ca91cx42_device;
/* Enable the device */
retval = pci_enable_device(pdev);
if (retval) {
dev_err(&pdev->dev, "Unable to enable device\n");
goto err_enable;
}
/* Map Registers */
retval = pci_request_regions(pdev, driver_name);
if (retval) {
dev_err(&pdev->dev, "Unable to reserve resources\n");
goto err_resource;
}
/* map registers in BAR 0 */
ca91cx42_device->base = ioremap_nocache(pci_resource_start(pdev, 0),
4096);
if (!ca91cx42_device->base) {
dev_err(&pdev->dev, "Unable to remap CRG region\n");
retval = -EIO;
goto err_remap;
}
/* Check to see if the mapping worked out */
data = ioread32(ca91cx42_device->base + CA91CX42_PCI_ID) & 0x0000FFFF;
if (data != PCI_VENDOR_ID_TUNDRA) {
dev_err(&pdev->dev, "PCI_ID check failed\n");
retval = -EIO;
goto err_test;
}
/* Initialize wait queues & mutual exclusion flags */
init_waitqueue_head(&ca91cx42_device->dma_queue);
init_waitqueue_head(&ca91cx42_device->iack_queue);
mutex_init(&ca91cx42_device->vme_int);
mutex_init(&ca91cx42_device->vme_rmw);
ca91cx42_bridge->parent = &pdev->dev;
strcpy(ca91cx42_bridge->name, driver_name);
/* Setup IRQ */
retval = ca91cx42_irq_init(ca91cx42_bridge);
if (retval != 0) {
dev_err(&pdev->dev, "Chip Initialization failed.\n");
goto err_irq;
}
/* Add master windows to list */
for (i = 0; i < CA91C142_MAX_MASTER; i++) {
master_image = kmalloc(sizeof(*master_image), GFP_KERNEL);
if (!master_image) {
retval = -ENOMEM;
goto err_master;
}
master_image->parent = ca91cx42_bridge;
spin_lock_init(&master_image->lock);
master_image->locked = 0;
master_image->number = i;
master_image->address_attr = VME_A16 | VME_A24 | VME_A32 |
VME_CRCSR | VME_USER1 | VME_USER2;
master_image->cycle_attr = VME_SCT | VME_BLT | VME_MBLT |
VME_SUPER | VME_USER | VME_PROG | VME_DATA;
master_image->width_attr = VME_D8 | VME_D16 | VME_D32 | VME_D64;
memset(&master_image->bus_resource, 0,
sizeof(master_image->bus_resource));
master_image->kern_base = NULL;
list_add_tail(&master_image->list,
&ca91cx42_bridge->master_resources);
}
/* Add slave windows to list */
for (i = 0; i < CA91C142_MAX_SLAVE; i++) {
slave_image = kmalloc(sizeof(*slave_image), GFP_KERNEL);
if (!slave_image) {
retval = -ENOMEM;
goto err_slave;
}
slave_image->parent = ca91cx42_bridge;
mutex_init(&slave_image->mtx);
slave_image->locked = 0;
slave_image->number = i;
slave_image->address_attr = VME_A24 | VME_A32 | VME_USER1 |
VME_USER2;
/* Only windows 0 and 4 support A16 */
if (i == 0 || i == 4)
slave_image->address_attr |= VME_A16;
slave_image->cycle_attr = VME_SCT | VME_BLT | VME_MBLT |
VME_SUPER | VME_USER | VME_PROG | VME_DATA;
list_add_tail(&slave_image->list,
&ca91cx42_bridge->slave_resources);
}
/* Add dma engines to list */
for (i = 0; i < CA91C142_MAX_DMA; i++) {
dma_ctrlr = kmalloc(sizeof(*dma_ctrlr), GFP_KERNEL);
if (!dma_ctrlr) {
retval = -ENOMEM;
goto err_dma;
}
dma_ctrlr->parent = ca91cx42_bridge;
mutex_init(&dma_ctrlr->mtx);
dma_ctrlr->locked = 0;
dma_ctrlr->number = i;
dma_ctrlr->route_attr = VME_DMA_VME_TO_MEM |
VME_DMA_MEM_TO_VME;
INIT_LIST_HEAD(&dma_ctrlr->pending);
INIT_LIST_HEAD(&dma_ctrlr->running);
list_add_tail(&dma_ctrlr->list,
&ca91cx42_bridge->dma_resources);
}
/* Add location monitor to list */
lm = kmalloc(sizeof(*lm), GFP_KERNEL);
if (!lm) {
retval = -ENOMEM;
goto err_lm;
}
lm->parent = ca91cx42_bridge;
mutex_init(&lm->mtx);
lm->locked = 0;
lm->number = 1;
lm->monitors = 4;
list_add_tail(&lm->list, &ca91cx42_bridge->lm_resources);
ca91cx42_bridge->slave_get = ca91cx42_slave_get;
ca91cx42_bridge->slave_set = ca91cx42_slave_set;
ca91cx42_bridge->master_get = ca91cx42_master_get;
ca91cx42_bridge->master_set = ca91cx42_master_set;
ca91cx42_bridge->master_read = ca91cx42_master_read;
ca91cx42_bridge->master_write = ca91cx42_master_write;
ca91cx42_bridge->master_rmw = ca91cx42_master_rmw;
ca91cx42_bridge->dma_list_add = ca91cx42_dma_list_add;
ca91cx42_bridge->dma_list_exec = ca91cx42_dma_list_exec;
ca91cx42_bridge->dma_list_empty = ca91cx42_dma_list_empty;
ca91cx42_bridge->irq_set = ca91cx42_irq_set;
ca91cx42_bridge->irq_generate = ca91cx42_irq_generate;
ca91cx42_bridge->lm_set = ca91cx42_lm_set;
ca91cx42_bridge->lm_get = ca91cx42_lm_get;
ca91cx42_bridge->lm_attach = ca91cx42_lm_attach;
ca91cx42_bridge->lm_detach = ca91cx42_lm_detach;
ca91cx42_bridge->slot_get = ca91cx42_slot_get;
ca91cx42_bridge->alloc_consistent = ca91cx42_alloc_consistent;
ca91cx42_bridge->free_consistent = ca91cx42_free_consistent;
data = ioread32(ca91cx42_device->base + MISC_CTL);
dev_info(&pdev->dev, "Board is%s the VME system controller\n",
(data & CA91CX42_MISC_CTL_SYSCON) ? "" : " not");
dev_info(&pdev->dev, "Slot ID is %d\n",
ca91cx42_slot_get(ca91cx42_bridge));
if (ca91cx42_crcsr_init(ca91cx42_bridge, pdev))
dev_err(&pdev->dev, "CR/CSR configuration failed.\n");
/* Need to save ca91cx42_bridge pointer locally in link list for use in
* ca91cx42_remove()
*/
retval = vme_register_bridge(ca91cx42_bridge);
if (retval != 0) {
dev_err(&pdev->dev, "Chip Registration failed.\n");
goto err_reg;
}
pci_set_drvdata(pdev, ca91cx42_bridge);
return 0;
err_reg:
ca91cx42_crcsr_exit(ca91cx42_bridge, pdev);
err_lm:
/* resources are stored in link list */
list_for_each_safe(pos, n, &ca91cx42_bridge->lm_resources) {
lm = list_entry(pos, struct vme_lm_resource, list);
list_del(pos);
kfree(lm);
}
err_dma:
/* resources are stored in link list */
list_for_each_safe(pos, n, &ca91cx42_bridge->dma_resources) {
dma_ctrlr = list_entry(pos, struct vme_dma_resource, list);
list_del(pos);
kfree(dma_ctrlr);
}
err_slave:
/* resources are stored in link list */
list_for_each_safe(pos, n, &ca91cx42_bridge->slave_resources) {
slave_image = list_entry(pos, struct vme_slave_resource, list);
list_del(pos);
kfree(slave_image);
}
err_master:
/* resources are stored in link list */
list_for_each_safe(pos, n, &ca91cx42_bridge->master_resources) {
master_image = list_entry(pos, struct vme_master_resource,
list);
list_del(pos);
kfree(master_image);
}
ca91cx42_irq_exit(ca91cx42_device, pdev);
err_irq:
err_test:
iounmap(ca91cx42_device->base);
err_remap:
pci_release_regions(pdev);
err_resource:
pci_disable_device(pdev);
err_enable:
kfree(ca91cx42_device);
err_driver:
kfree(ca91cx42_bridge);
err_struct:
return retval;
}
static void ca91cx42_remove(struct pci_dev *pdev)
{
struct list_head *pos = NULL, *n;
struct vme_master_resource *master_image;
struct vme_slave_resource *slave_image;
struct vme_dma_resource *dma_ctrlr;
struct vme_lm_resource *lm;
struct ca91cx42_driver *bridge;
struct vme_bridge *ca91cx42_bridge = pci_get_drvdata(pdev);
bridge = ca91cx42_bridge->driver_priv;
/* Turn off Ints */
iowrite32(0, bridge->base + LINT_EN);
/* Turn off the windows */
iowrite32(0x00800000, bridge->base + LSI0_CTL);
iowrite32(0x00800000, bridge->base + LSI1_CTL);
iowrite32(0x00800000, bridge->base + LSI2_CTL);
iowrite32(0x00800000, bridge->base + LSI3_CTL);
iowrite32(0x00800000, bridge->base + LSI4_CTL);
iowrite32(0x00800000, bridge->base + LSI5_CTL);
iowrite32(0x00800000, bridge->base + LSI6_CTL);
iowrite32(0x00800000, bridge->base + LSI7_CTL);
iowrite32(0x00F00000, bridge->base + VSI0_CTL);
iowrite32(0x00F00000, bridge->base + VSI1_CTL);
iowrite32(0x00F00000, bridge->base + VSI2_CTL);
iowrite32(0x00F00000, bridge->base + VSI3_CTL);
iowrite32(0x00F00000, bridge->base + VSI4_CTL);
iowrite32(0x00F00000, bridge->base + VSI5_CTL);
iowrite32(0x00F00000, bridge->base + VSI6_CTL);
iowrite32(0x00F00000, bridge->base + VSI7_CTL);
vme_unregister_bridge(ca91cx42_bridge);
ca91cx42_crcsr_exit(ca91cx42_bridge, pdev);
/* resources are stored in link list */
list_for_each_safe(pos, n, &ca91cx42_bridge->lm_resources) {
lm = list_entry(pos, struct vme_lm_resource, list);
list_del(pos);
kfree(lm);
}
/* resources are stored in link list */
list_for_each_safe(pos, n, &ca91cx42_bridge->dma_resources) {
dma_ctrlr = list_entry(pos, struct vme_dma_resource, list);
list_del(pos);
kfree(dma_ctrlr);
}
/* resources are stored in link list */
list_for_each_safe(pos, n, &ca91cx42_bridge->slave_resources) {
slave_image = list_entry(pos, struct vme_slave_resource, list);
list_del(pos);
kfree(slave_image);
}
/* resources are stored in link list */
list_for_each_safe(pos, n, &ca91cx42_bridge->master_resources) {
master_image = list_entry(pos, struct vme_master_resource,
list);
list_del(pos);
kfree(master_image);
}
ca91cx42_irq_exit(bridge, pdev);
iounmap(bridge->base);
pci_release_regions(pdev);
pci_disable_device(pdev);
kfree(ca91cx42_bridge);
}
module_pci_driver(ca91cx42_driver);
MODULE_PARM_DESC(geoid, "Override geographical addressing");
module_param(geoid, int, 0);
MODULE_DESCRIPTION("VME driver for the Tundra Universe II VME bridge");
MODULE_LICENSE("GPL");