On Wed, Apr 24, 2013 at 01:25:08AM +1000, Benjamin Herrenschmidt wrote:
>On Tue, 2013-04-23 at 19:03 +0800, Gavin Shan wrote:
>> * of flags if that becomes the case
>> */
>> if (tbl->it_type & TCE_PCI_SWINV_CREATE)
>> - pnv_tce_invalidate(tbl, tces, tcep - 1)
On Tue, 2013-04-23 at 19:03 +0800, Gavin Shan wrote:
> * of flags if that becomes the case
> */
> if (tbl->it_type & TCE_PCI_SWINV_CREATE)
> - pnv_tce_invalidate(tbl, tces, tcep - 1);
> + phb->dma_tce_invalidate(tbl, tces, tcep - 1);
>
>
The TCE should be invalidated while it's created or free'd. The
approach to do that for IODA1 and IODA2 compliant PHBs are different.
So the patch differentiate them with virtualized interface hooked
to the PHB. It's notable that the PCI address is used to invalidate
the corresponding TCE on IODA2