* [PATCH 2.6] Cobalt 1/2: tidy up PCI fixups
@ 2005-03-01 8:38 Peter Horton
2005-03-04 14:38 ` Ralf Baechle
0 siblings, 1 reply; 2+ messages in thread
From: Peter Horton @ 2005-03-01 8:38 UTC (permalink / raw)
To: linux-mips
It's not necessary to hide Galileo if we correctly mark it as a host
bridge.
P.
--
diff -urpN linux-cvs/arch/mips/cobalt/setup.c linux-wip/arch/mips/cobalt/setup.c
--- linux-cvs/arch/mips/cobalt/setup.c 2005-02-21 16:18:36.000000000 +0000
+++ linux-wip/arch/mips/cobalt/setup.c 2005-02-27 11:37:10.000000000 +0000
@@ -62,11 +62,11 @@ static void __init cobalt_timer_setup(st
extern struct pci_ops gt64111_pci_ops;
static struct resource cobalt_mem_resource = {
- "GT64111 PCI MEM", GT64111_MEM_BASE, GT64111_MEM_END, IORESOURCE_MEM
+ "PCI memory", GT64111_MEM_BASE, GT64111_MEM_END, IORESOURCE_MEM
};
static struct resource cobalt_io_resource = {
- "GT64111 IO MEM", 0x00001000UL, GT64111_IO_END - GT64111_IO_BASE, IORESOURCE_IO
+ "PCI I/O", 0x1000, 0xffff, IORESOURCE_IO
};
static struct resource cobalt_io_resources[] = {
@@ -100,7 +100,7 @@ static int __init cobalt_setup(void)
set_io_port_base(CKSEG1ADDR(GT64111_IO_BASE));
- /* IO region should cover all Galileo IO */
+ /* I/O port resource must include UART and LCD/buttons */
ioport_resource.end = 0x0fffffff;
/*
diff -urpN linux-cvs/arch/mips/pci/fixup-cobalt.c linux-wip/arch/mips/pci/fixup-cobalt.c
--- linux-cvs/arch/mips/pci/fixup-cobalt.c 2005-02-21 16:24:02.000000000 +0000
+++ linux-wip/arch/mips/pci/fixup-cobalt.c 2005-02-27 11:24:44.000000000 +0000
@@ -21,6 +21,20 @@
extern int cobalt_board_id;
+static void qube_raq_galileo_early_fixup(struct pci_dev *dev)
+{
+ if (dev->devfn == PCI_DEVFN(0, 0) &&
+ (dev->class >> 8) == PCI_CLASS_MEMORY_OTHER) {
+
+ dev->class = (PCI_CLASS_BRIDGE_HOST << 8) | (dev->class & 0xff);
+
+ printk(KERN_INFO "Galileo: fixed bridge class\n");
+ }
+}
+
+DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_MARVELL, PCI_DEVICE_ID_MARVELL_GT64111,
+ qube_raq_galileo_early_fixup);
+
static void qube_raq_via_bmIDE_fixup(struct pci_dev *dev)
{
unsigned short cfgword;
@@ -47,8 +61,10 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_V
static void qube_raq_galileo_fixup(struct pci_dev *dev)
{
unsigned short galileo_id;
- int i;
+ if (dev->devfn != PCI_DEVFN(0, 0))
+ return;
+
/* Fix PCI latency-timer and cache-line-size values in Galileo
* host bridge.
*/
@@ -73,7 +89,7 @@ static void qube_raq_galileo_fixup(struc
pci_read_config_word(dev, PCI_REVISION_ID, &galileo_id);
galileo_id &= 0xff; /* mask off class info */
- printk("Galileo ID: %u\n", galileo_id);
+ printk(KERN_INFO "Galileo: revision %u\n", galileo_id);
#if 0
if (galileo_id >= 0x10) {
@@ -88,18 +104,9 @@ static void qube_raq_galileo_fixup(struc
/* Old Galileo, assumes PCI STOP line to VIA is disconnected. */
GALILEO_OUTL(0xffff, GT_PCI0_TOR_OFS);
}
-
- /*
- * hide Galileo from the kernel's PCI resource assignment. The BARs
- * on Galileo will already have been set up by the boot loader to
- * match the DRAM configuration so we don't want them being monkeyed
- * around with.
- */
- for (i = 0; i < DEVICE_COUNT_RESOURCE; ++i)
- dev->resource[i].start = dev->resource[i].end = dev->resource[i].flags = 0;
}
-DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL, PCI_ANY_ID,
+DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_MARVELL, PCI_DEVICE_ID_MARVELL_GT64111,
qube_raq_galileo_fixup);
static char irq_tab_cobalt[] __initdata = {
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH 2.6] Cobalt 1/2: tidy up PCI fixups
2005-03-01 8:38 [PATCH 2.6] Cobalt 1/2: tidy up PCI fixups Peter Horton
@ 2005-03-04 14:38 ` Ralf Baechle
0 siblings, 0 replies; 2+ messages in thread
From: Ralf Baechle @ 2005-03-04 14:38 UTC (permalink / raw)
To: Peter Horton; +Cc: linux-mips
On Tue, Mar 01, 2005 at 08:38:52AM +0000, Peter Horton wrote:
> It's not necessary to hide Galileo if we correctly mark it as a host
> bridge.
Indeed, that's the right thing to do and it should be done on a bunch
of other systems also.
Ralf
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2005-03-04 14:39 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-03-01 8:38 [PATCH 2.6] Cobalt 1/2: tidy up PCI fixups Peter Horton
2005-03-04 14:38 ` Ralf Baechle
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox