On 08/19/2014 10:43 AM, David Gibson wrote:
> On Fri, Aug 15, 2014 at 08:12:24PM +1000, Alexey Kardashevskiy wrote:
>> The existing KVM_CREATE_SPAPR_TCE ioctl only support 4G windows max.
>> We are going to add huge DMA windows support so this will create small
>> window and unexpectedly fail later.
>>
>> This disables KVM_CREATE_SPAPR_TCE for windows bigger that 4GB. Since
>> those windows are normally mapped at the boot time, there will be no
>> performance impact.
>>
>> Signed-off-by: Alexey Kardashevskiy <a...@ozlabs.ru>
>
> I think perhaps the crucial point here is that the window size
> parameter to the kernel ioctl() is only 32-bit, so there's no way of
> expressing a TCE window > 4GB.
That exactly the point. I'll steal these bit and commit log will start with:
===
The existing KVM_CREATE_SPAPR_TCE ioctl only support 4G windows max a
the window size parameter to the kernel ioctl() is only 32-bit, so
there's no way of expressing a TCE window > 4GB.
===
>> ---
>> hw/ppc/spapr_iommu.c | 6 +++---
>> 1 file changed, 3 insertions(+), 3 deletions(-)
>>
>> diff --git a/hw/ppc/spapr_iommu.c b/hw/ppc/spapr_iommu.c
>> index f6e32a4..36f5d27 100644
>> --- a/hw/ppc/spapr_iommu.c
>> +++ b/hw/ppc/spapr_iommu.c
>> @@ -113,11 +113,11 @@ static MemoryRegionIOMMUOps spapr_iommu_ops = {
>> static int spapr_tce_table_realize(DeviceState *dev)
>> {
>> sPAPRTCETable *tcet = SPAPR_TCE_TABLE(dev);
>> + uint64_t window_size = tcet->nb_table << tcet->page_shift;
>
> tcet->nb_table is only 32-bit itself, so this isn't going to work as
> intended without a cast.
Oh. Thanks. /me is thinking how to catch this kind of errors from a script...
--
Alexey
