>No, just map an uncached, unbuffered memmory space for the PCI environment >and the problem will go away (if it is a cache related issue, that is...) It's not quite that simple. We already map PCI I/O and memory uncached. The difficulty with DMA-enabled network drivers is that they want to access main memory which obviously needs to be cached for performance reasons. So explicit synchronisation instructions are needed. We've previously kicked around the idea of adding a GFP_UNCACHED flag to __get_free_page to help with this but nobody has implemented it yet. p. unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]
- StrongARM and AMD LANCE PCI NETCARD christophe leroy
- Re: StrongARM and AMD LANCE PCI NETCARD Philip Blundell
- Re: StrongARM and AMD LANCE PCI NETCARD Dr . David Alan Gilbert
- Re: StrongARM and AMD LANCE PCI NETCARD Russell King
- Re: StrongARM and AMD LANCE PCI NETCA... christophe leroy
- Re: StrongARM and AMD LANCE PCI N... Russell King
- Re: StrongARM and AMD LANCE PCI NETCA... Philip Blundell
- Re: StrongARM and AMD LANCE PCI NETCARD Conor O'Gorman
- Re: StrongARM and AMD LANCE PCI NETCA... Philip Blundell
- Re: StrongARM and AMD LANCE PCI NETCA... Olof Backing
- Re: StrongARM and AMD LANCE PCI N... Philip Blundell
- Re: StrongARM and AMD LANCE ... Alan Cox
- Re: StrongARM and AMD LA... Olof Backing
- Re: StrongARM and AMD LA... Philip Blundell
- Re: StrongARM and AM... Paul Koning
- Re: StrongARM and AM... Russell King
- Re: StrongARM and AMD LANCE PCI N... Dr . David Alan Gilbert
