lkml.org 
[lkml]   [2009]   [Mar]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [PATCH] pci: don't enable too many HT MSI mapping
Date
On Samstag 28 Februar 2009 23:43:51 Yinghai Lu wrote:
> please check this one
> v1 is pulled by Linus from Jesse's pci tree already, and v3 is needed
> to be applied on top of v1
>
> YH

Hi,

yes, this works nicely now. Thanks. (Are you still interested in all the infos
you requested in your last email?) But have you looked at the original patch?
It might be that your current one is not enough to cover all models. In the
original NVidia one following bridges were enabled:

+#define PCI_DEVICE_ID_NVIDIA_NFORCE_C51_MEMC0 0x02F0
+#define PCI_DEVICE_ID_NVIDIA_NFORCE_C51_MEMC1 0x02F1
+#define PCI_DEVICE_ID_NVIDIA_NFORCE_C51_MEMC2 0x02F2
+#define PCI_DEVICE_ID_NVIDIA_NFORCE_C51_MEMC3 0x02F3
+#define PCI_DEVICE_ID_NVIDIA_NFORCE_C51_MEMC4 0x02F4
+#define PCI_DEVICE_ID_NVIDIA_NFORCE_C51_MEMC5 0x02F5
+#define PCI_DEVICE_ID_NVIDIA_NFORCE_C51_MEMC6 0x02F6
+#define PCI_DEVICE_ID_NVIDIA_NFORCE_C51_MEMC7 0x02F7
+#define PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_MEMC 0x0369

As you can see, the one on my board is the MEMC0. I don't know which revisions
have the other ones. The MEMC case had a special bahaviour in the original
patch also:

+ bridge_dev = NULL;
+ while ((bridge_dev = pci_get_device(PCI_VENDOR_ID_NVIDIA,
+ PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_MEMC, bridge_dev))
+ != NULL) {
+ pci_enable_msi_ht_cap(bridge_dev);
+ }

Cheers,

Prakash


\
 
 \ /
  Last update: 2009-03-01 08:51    [W:0.096 / U:0.500 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site