[PATCH]fix for Intel 8xx agp in 2.4.15-pre5

[Nicolas Aspert <Nicolas.Aspert@epfl.ch>]

[-- Attachment #1: Type: text/plain, Size: 308 bytes --]

Hello

There is a mistake in the routine 'intel_8xx_tlbflush' in the agpgart 
module (and I think I am responsible for it... shame on me !).
Anyway here is the patch (against 2.4.15-pre5).

Best regards.
-- 
Nicolas Aspert      Signal Processing Laboratory (LTS)
Swiss Federal Institute of Technology (EPFL)

[-- Attachment #2: patch-agp_intel_8xx-2.4.15-pre5_fix_tlbflush --]
[-- Type: text/plain, Size: 655 bytes --]

diff -Nru linux-2.4.15-pre5.clean/drivers/char/agp/agpgart_be.c linux-2.4.15-pre5/drivers/char/agp/agpgart_be.c
--- linux-2.4.15-pre5.clean/drivers/char/agp/agpgart_be.c	Fri Nov 16 08:45:44 2001
+++ linux-2.4.15-pre5/drivers/char/agp/agpgart_be.c	Fri Nov 16 13:51:44 2001
@@ -1462,7 +1462,7 @@
   pci_read_config_dword(agp_bridge.dev, INTEL_AGPCTRL, &temp);
   pci_write_config_dword(agp_bridge.dev, INTEL_AGPCTRL, temp & ~(1 << 7));
   pci_read_config_dword(agp_bridge.dev, INTEL_AGPCTRL, &temp);
-  pci_write_config_dword(agp_bridge.dev, INTEL_AGPCTRL, temp & (1 << 7));
+  pci_write_config_dword(agp_bridge.dev, INTEL_AGPCTRL, temp | (1 << 7));
 }