Skip to contentSkip to conversation

Map the graphic card to the bridge its connected to.

AGPGART Benjamin Herrenschmidt 1 participant 2 messages Mar 10, 2005 – Mar 11, 2005 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw
Quoted 16 lines · Linux Kernel Mailing List

ChangeSet 1.1982.82.19, 2005/02/22 21:25:33-05:00, [email protected]

[AGPGART] Map the graphic card to the bridge its connected to.

 generic.c |    5 +++++
 1 files changed, 5 insertions(+)

diff -Nru a/drivers/char/agp/generic.c b/drivers/char/agp/generic.c

drivers/char/agp/generic.c line 645

Quoted 10 lines
drivers/char/agp/generic.c +5 -0
@@ -636,6 +636,11 @@
 			pci_dev_put(device);
 			continue;
 		}
+		if ((device->bus->self->vendor != bridge->dev->vendor) &&
+			(device->bus->self->device != bridge->dev->device)) {
+			pci_dev_put(device);
+			continue;
+		}
 	}
 

That sounds totally bogus and blows up on pmac, please revert.

device->bus may be a host bridge, which has no bus->self -> Ooops.

Unfortunately, there is no sane way to match a host bridge with it's eventual "self" device if it has any. The only way would be to scan for devices of class host bridge, but that isn't even 100% reliable.

The result is that the self device (AGP bridge device) is generally a sibling of the actual AGP card, which is source of interesting problems, especially with power management.

Ideally, the AGP property should be implemented at the "bus instance" level, but we don't really have a real pci bus driver layer at this point, and we can't use normal PCI discovery to find host bridges, so that would require arch support.

In the meantime, please revert the above, it will just blow up on a number of setups.

Ben.

#2 lore raw
Quoted 6 lines · Benjamin Herrenschmidt
Quoted 5 lines · Linux Kernel Mailing List

ChangeSet 1.1982.82.19, 2005/02/22 21:25:33-05:00, [email protected]

[AGPGART] Map the graphic card to the bridge its connected to.

Note that the effect of reverting the patch gets a broken result too (btw... the code had a spurrious pci_dev_put() too ...)

Anyway, paulus is working on a solution that should be good enough for both cases where the AGP bridge is a sibling of the device, and where the bridge is a parent of the device. Patch soon...

Ben.