Skip to contentSkip to conversation

Assign all PCI busses on G3 Blue & White

PATCH In review Pavel Roskin 2 participants 2 messages Feb 4, 2007 – Feb 5, 2007 PowerPC

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Assign all PCI busses on G3 Blue & White

From: Pavel Roskin <[email protected]>

G3 Blue & White is misconfigured by default so that CardBus controllers in PCI slots don't work. The PCI bridge is programmed to only allow access to bus 1 but not higher busses.

The patch forces the PCI busses to be reassigned if a Grackle controller is found and the machine identifies itself as "PowerMac1,1"

---
My previous attempt to fix this bug can be found in this thread:
http://ozlabs.org/pipermail/linuxppc-dev/2004-May/thread.html#17056
The new patch relies on the existing common PCI code and doesn't poke bridge registers itself. It would even do the right thing if a PCI or CardBus bridge is inserted into the upper slot reserved for video cards and connected directly to bus 0.
---

 arch/powerpc/sysdev/grackle.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)
arch/powerpc/sysdev/grackle.c +2 -0
index b6ec793..4205362 100644
@@ -56,6 +56,8 @@ static inline void grackle_set_loop_snoop(struct pci_controller *bp, int enable)
 void __init setup_grackle(struct pci_controller *hose)
 {
 	setup_indirect_pci(hose, 0xfec00000, 0xfee00000);
+	if (machine_is_compatible("PowerMac1,1"))
+		pci_assign_all_buses = 1;
 	if (machine_is_compatible("AAPL,PowerBook1998"))
 		grackle_set_loop_snoop(hose, 1);
 #if 0	/* Disabled for now, HW problems ??? */
Signature · 2 lines
Regards,
Pavel Roskin
Acked-by on v1 1/1
#1 lore raw
Quoted 10 lines · Pavel Roskin

From: Pavel Roskin <[email protected]>

G3 Blue & White is misconfigured by default so that CardBus controllers in PCI slots don't work. The PCI bridge is programmed to only allow access to bus 1 but not higher busses.

The patch forces the PCI busses to be reassigned if a Grackle controller is found and the machine identifies itself as "PowerMac1,1"

Quoted 27 lines

---
My previous attempt to fix this bug can be found in this thread:
http://ozlabs.org/pipermail/linuxppc-dev/2004-May/thread.html#17056
The new patch relies on the existing common PCI code and doesn't poke bridge registers itself. It would even do the right thing if a PCI or CardBus bridge is inserted into the upper slot reserved for video cards and connected directly to bus 0.
---

 arch/powerpc/sysdev/grackle.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)
arch/powerpc/sysdev/grackle.c +2 -0
index b6ec793..4205362 100644
@@ -56,6 +56,8 @@ static inline void grackle_set_loop_snoop(struct pci_controller *bp, int enable)
 void __init setup_grackle(struct pci_controller *hose)
 {
 	setup_indirect_pci(hose, 0xfec00000, 0xfee00000);
+	if (machine_is_compatible("PowerMac1,1"))
+		pci_assign_all_buses = 1;
 	if (machine_is_compatible("AAPL,PowerBook1998"))
 		grackle_set_loop_snoop(hose, 1);
 #if 0	/* Disabled for now, HW problems ??? */