summaryrefslogtreecommitdiffstats
path: root/drivers/macintosh/macio_asic.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/macintosh/macio_asic.c')
-rw-r--r--drivers/macintosh/macio_asic.c254
1 files changed, 148 insertions, 106 deletions
diff --git a/drivers/macintosh/macio_asic.c b/drivers/macintosh/macio_asic.c
index 228e185..2a545ce 100644
--- a/drivers/macintosh/macio_asic.c
+++ b/drivers/macintosh/macio_asic.c
@@ -3,6 +3,13 @@
* a MacIO ASIC. Interface to new driver model mostly
* stolen from the PCI version.
*
+ * Copyright (C) 2005 Ben. Herrenschmidt (benh@kernel.crashing.org)
+ *
+ * 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.
+ *
* TODO:
*
* - Don't probe below media bay by default, but instead provide
@@ -218,12 +225,14 @@ postcore_initcall(macio_bus_driver_init);
/**
- * macio_release_dev - free a macio device structure when all users of it are finished.
+ * macio_release_dev - free a macio device structure when all users of it are
+ * finished.
* @dev: device that's been disconnected
*
- * Will be called only by the device core when all users of this macio device are
- * done. This currently means never as we don't hot remove any macio device yet,
- * though that will happen with mediabay based devices in a later implementation.
+ * Will be called only by the device core when all users of this macio device
+ * are done. This currently means never as we don't hot remove any macio
+ * device yet, though that will happen with mediabay based devices in a later
+ * implementation.
*/
static void macio_release_dev(struct device *dev)
{
@@ -242,49 +251,114 @@ static void macio_release_dev(struct device *dev)
* If this routine returns non-null, then the resource is completely
* skipped.
*/
-static int macio_resource_quirks(struct device_node *np, struct resource *res, int index)
+static int macio_resource_quirks(struct device_node *np, struct resource *res,
+ int index)
{
if (res->flags & IORESOURCE_MEM) {
/* Grand Central has too large resource 0 on some machines */
- if (index == 0 && !strcmp(np->name, "gc")) {
- np->addrs[0].size = 0x20000;
+ if (index == 0 && !strcmp(np->name, "gc"))
res->end = res->start + 0x1ffff;
- }
+
/* Airport has bogus resource 2 */
if (index >= 2 && !strcmp(np->name, "radio"))
return 1;
+
+#ifndef CONFIG_PPC64
/* DBDMAs may have bogus sizes */
- if ((res->start & 0x0001f000) == 0x00008000) {
- np->addrs[index].size = 0x100;
+ if ((res->start & 0x0001f000) == 0x00008000)
res->end = res->start + 0xff;
- }
- /* ESCC parent eats child resources. We could have added a level of hierarchy,
- * but I don't really feel the need for it */
+#endif /* CONFIG_PPC64 */
+
+ /* ESCC parent eats child resources. We could have added a
+ * level of hierarchy, but I don't really feel the need
+ * for it
+ */
if (!strcmp(np->name, "escc"))
return 1;
+
/* ESCC has bogus resources >= 3 */
- if (index >= 3 && !(strcmp(np->name, "ch-a") && strcmp(np->name, "ch-b")))
+ if (index >= 3 && !(strcmp(np->name, "ch-a") &&
+ strcmp(np->name, "ch-b")))
return 1;
+
/* Media bay has too many resources, keep only first one */
if (index > 0 && !strcmp(np->name, "media-bay"))
return 1;
+
/* Some older IDE resources have bogus sizes */
if (!(strcmp(np->name, "IDE") && strcmp(np->name, "ATA") &&
strcmp(np->type, "ide") && strcmp(np->type, "ata"))) {
- if (index == 0 && np->addrs[0].size > 0x1000) {
- np->addrs[0].size = 0x1000;
+ if (index == 0 && (res->end - res->start) > 0xfff)
res->end = res->start + 0xfff;
- }
- if (index == 1 && np->addrs[1].size > 0x100) {
- np->addrs[1].size = 0x100;
+ if (index == 1 && (res->end - res->start) > 0xff)
res->end = res->start + 0xff;
- }
}
}
return 0;
}
+static void macio_setup_interrupts(struct macio_dev *dev)
+{
+ struct device_node *np = dev->ofdev.node;
+ int i,j;
+
+ /* For now, we use pre-parsed entries in the device-tree for
+ * interrupt routing and addresses, but we should change that
+ * to dynamically parsed entries and so get rid of most of the
+ * clutter in struct device_node
+ */
+ for (i = j = 0; i < np->n_intrs; i++) {
+ struct resource *res = &dev->interrupt[j];
+
+ if (j >= MACIO_DEV_COUNT_IRQS)
+ break;
+ res->start = np->intrs[i].line;
+ res->flags = IORESOURCE_IO;
+ if (np->intrs[j].sense)
+ res->flags |= IORESOURCE_IRQ_LOWLEVEL;
+ else
+ res->flags |= IORESOURCE_IRQ_HIGHEDGE;
+ res->name = dev->ofdev.dev.bus_id;
+ if (macio_resource_quirks(np, res, i))
+ memset(res, 0, sizeof(struct resource));
+ else
+ j++;
+ }
+ dev->n_interrupts = j;
+}
+
+static void macio_setup_resources(struct macio_dev *dev,
+ struct resource *parent_res)
+{
+ struct device_node *np = dev->ofdev.node;
+ struct resource r;
+ int index;
+
+ for (index = 0; of_address_to_resource(np, index, &r) == 0; index++) {
+ struct resource *res = &dev->resource[index];
+ if (index >= MACIO_DEV_COUNT_RESOURCES)
+ break;
+ *res = r;
+ res->name = dev->ofdev.dev.bus_id;
+
+ if (macio_resource_quirks(np, res, index)) {
+ memset(res, 0, sizeof(struct resource));
+ continue;
+ }
+ /* Currently, we consider failure as harmless, this may
+ * change in the future, once I've found all the device
+ * tree bugs in older machines & worked around them
+ */
+ if (insert_resource(parent_res, res)) {
+ printk(KERN_WARNING "Can't request resource "
+ "%d for MacIO device %s\n",
+ index, dev->ofdev.dev.bus_id);
+ }
+ }
+ dev->n_resources = index;
+}
+
/**
* macio_add_one_device - Add one device from OF node to the device tree
* @chip: pointer to the macio_chip holding the device
@@ -294,12 +368,13 @@ static int macio_resource_quirks(struct device_node *np, struct resource *res, i
* When media-bay is changed to hotswap drivers, this function will
* be exposed to the bay driver some way...
*/
-static struct macio_dev * macio_add_one_device(struct macio_chip *chip, struct device *parent,
- struct device_node *np, struct macio_dev *in_bay,
+static struct macio_dev * macio_add_one_device(struct macio_chip *chip,
+ struct device *parent,
+ struct device_node *np,
+ struct macio_dev *in_bay,
struct resource *parent_res)
{
struct macio_dev *dev;
- int i, j;
u32 *reg;
if (np == NULL)
@@ -326,7 +401,8 @@ static struct macio_dev * macio_add_one_device(struct macio_chip *chip, struct d
/* MacIO itself has a different reg, we use it's PCI base */
if (np == chip->of_node) {
- sprintf(dev->ofdev.dev.bus_id, "%1d.%08lx:%.*s", chip->lbus.index,
+ sprintf(dev->ofdev.dev.bus_id, "%1d.%08lx:%.*s",
+ chip->lbus.index,
#ifdef CONFIG_PCI
pci_resource_start(chip->lbus.pdev, 0),
#else
@@ -335,57 +411,16 @@ static struct macio_dev * macio_add_one_device(struct macio_chip *chip, struct d
MAX_NODE_NAME_SIZE, np->name);
} else {
reg = (u32 *)get_property(np, "reg", NULL);
- sprintf(dev->ofdev.dev.bus_id, "%1d.%08x:%.*s", chip->lbus.index,
+ sprintf(dev->ofdev.dev.bus_id, "%1d.%08x:%.*s",
+ chip->lbus.index,
reg ? *reg : 0, MAX_NODE_NAME_SIZE, np->name);
}
- /* For now, we use pre-parsed entries in the device-tree for
- * interrupt routing and addresses, but we should change that
- * to dynamically parsed entries and so get rid of most of the
- * clutter in struct device_node
- */
- for (i = j = 0; i < np->n_intrs; i++) {
- struct resource *res = &dev->interrupt[j];
-
- if (j >= MACIO_DEV_COUNT_IRQS)
- break;
- res->start = np->intrs[i].line;
- res->flags = IORESOURCE_IO;
- if (np->intrs[j].sense)
- res->flags |= IORESOURCE_IRQ_LOWLEVEL;
- else
- res->flags |= IORESOURCE_IRQ_HIGHEDGE;
- res->name = dev->ofdev.dev.bus_id;
- if (macio_resource_quirks(np, res, i))
- memset(res, 0, sizeof(struct resource));
- else
- j++;
- }
- dev->n_interrupts = j;
- for (i = j = 0; i < np->n_addrs; i++) {
- struct resource *res = &dev->resource[j];
-
- if (j >= MACIO_DEV_COUNT_RESOURCES)
- break;
- res->start = np->addrs[i].address;
- res->end = np->addrs[i].address + np->addrs[i].size - 1;
- res->flags = IORESOURCE_MEM;
- res->name = dev->ofdev.dev.bus_id;
- if (macio_resource_quirks(np, res, i))
- memset(res, 0, sizeof(struct resource));
- else {
- j++;
- /* Currently, we consider failure as harmless, this may
- * change in the future, once I've found all the device
- * tree bugs in older machines & worked around them
- */
- if (insert_resource(parent_res, res))
- printk(KERN_WARNING "Can't request resource %d for MacIO"
- " device %s\n", i, dev->ofdev.dev.bus_id);
- }
- }
- dev->n_resources = j;
+ /* Setup interrupts & resources */
+ macio_setup_interrupts(dev);
+ macio_setup_resources(dev, parent_res);
+ /* Register with core */
if (of_device_register(&dev->ofdev) != 0) {
printk(KERN_DEBUG"macio: device registration error for %s!\n",
dev->ofdev.dev.bus_id);
@@ -442,36 +477,42 @@ static void macio_pci_add_devices(struct macio_chip *chip)
/* First scan 1st level */
for (np = NULL; (np = of_get_next_child(pnode, np)) != NULL;) {
- if (!macio_skip_device(np)) {
- of_node_get(np);
- mdev = macio_add_one_device(chip, &rdev->ofdev.dev, np, NULL, root_res);
- if (mdev == NULL)
- of_node_put(np);
- else if (strncmp(np->name, "media-bay", 9) == 0)
- mbdev = mdev;
- else if (strncmp(np->name, "escc", 4) == 0)
- sdev = mdev;
- }
+ if (macio_skip_device(np))
+ continue;
+ of_node_get(np);
+ mdev = macio_add_one_device(chip, &rdev->ofdev.dev, np, NULL,
+ root_res);
+ if (mdev == NULL)
+ of_node_put(np);
+ else if (strncmp(np->name, "media-bay", 9) == 0)
+ mbdev = mdev;
+ else if (strncmp(np->name, "escc", 4) == 0)
+ sdev = mdev;
}
/* Add media bay devices if any */
if (mbdev)
- for (np = NULL; (np = of_get_next_child(mbdev->ofdev.node, np)) != NULL;)
- if (!macio_skip_device(np)) {
- of_node_get(np);
- if (macio_add_one_device(chip, &mbdev->ofdev.dev, np, mbdev,
- root_res) == NULL)
- of_node_put(np);
- }
+ for (np = NULL; (np = of_get_next_child(mbdev->ofdev.node, np))
+ != NULL;) {
+ if (macio_skip_device(np))
+ continue;
+ of_node_get(np);
+ if (macio_add_one_device(chip, &mbdev->ofdev.dev, np,
+ mbdev, root_res) == NULL)
+ of_node_put(np);
+ }
+
/* Add serial ports if any */
if (sdev) {
- for (np = NULL; (np = of_get_next_child(sdev->ofdev.node, np)) != NULL;)
- if (!macio_skip_device(np)) {
- of_node_get(np);
- if (macio_add_one_device(chip, &sdev->ofdev.dev, np, NULL,
- root_res) == NULL)
- of_node_put(np);
- }
+ for (np = NULL; (np = of_get_next_child(sdev->ofdev.node, np))
+ != NULL;) {
+ if (macio_skip_device(np))
+ continue;
+ of_node_get(np);
+ if (macio_add_one_device(chip, &sdev->ofdev.dev, np,
+ NULL, root_res) == NULL)
+ of_node_put(np);
+ }
}
}
@@ -519,7 +560,8 @@ void macio_unregister_driver(struct macio_driver *drv)
* Returns 0 on success, or %EBUSY on error. A warning
* message is also printed on failure.
*/
-int macio_request_resource(struct macio_dev *dev, int resource_no, const char *name)
+int macio_request_resource(struct macio_dev *dev, int resource_no,
+ const char *name)
{
if (macio_resource_len(dev, resource_no) == 0)
return 0;
@@ -606,20 +648,20 @@ static int __devinit macio_pci_probe(struct pci_dev *pdev, const struct pci_devi
if (ent->vendor != PCI_VENDOR_ID_APPLE)
return -ENODEV;
- /* Note regarding refcounting: We assume pci_device_to_OF_node() is ported
- * to new OF APIs and returns a node with refcount incremented. This isn't
- * the case today, but on the other hand ppc32 doesn't do refcounting. This
- * will have to be fixed when going to ppc64. --BenH.
+ /* Note regarding refcounting: We assume pci_device_to_OF_node() is
+ * ported to new OF APIs and returns a node with refcount incremented.
*/
np = pci_device_to_OF_node(pdev);
if (np == NULL)
return -ENODEV;
- /* This assumption is wrong, fix that here for now until I fix the arch */
+ /* The above assumption is wrong !!!
+ * fix that here for now until I fix the arch code
+ */
of_node_get(np);
- /* We also assume that pmac_feature will have done a get() on nodes stored
- * in the macio chips array
+ /* We also assume that pmac_feature will have done a get() on nodes
+ * stored in the macio chips array
*/
chip = macio_find(np, macio_unknown);
of_node_put(np);
@@ -639,9 +681,9 @@ static int __devinit macio_pci_probe(struct pci_dev *pdev, const struct pci_devi
/*
* HACK ALERT: The WallStreet PowerBook and some OHare based machines
- * have 2 macio ASICs. I must probe the "main" one first or IDE ordering
- * will be incorrect. So I put on "hold" the second one since it seem to
- * appear first on PCI
+ * have 2 macio ASICs. I must probe the "main" one first or IDE
+ * ordering will be incorrect. So I put on "hold" the second one since
+ * it seem to appear first on PCI
*/
if (chip->type == macio_gatwick || chip->type == macio_ohareII)
if (macio_chips[0].lbus.pdev == NULL) {
OpenPOWER on IntegriCloud