On 11/01/2015 11:53 AM, Thomas Monjalon wrote: > 2015-11-01 11:00, Marcel Apfelbaum: >> On 10/30/2015 07:48 PM, Thomas Monjalon wrote: >>> 2015-10-18 10:04, Michael S. Tsirkin: >>>> This will require more work then as 1.0 is a different >>>> endian-ness from 0.9. It's up to you guys to decide >>>> whether correct BE support is now a requirement for all >>>> new dpdk code. Let us know. >> >>> I'm not sure to understand. >>> Yes DPDK must work on big endian platforms. >>> Does this patch prevent from using virtio 0.9 with big endian? >> >> No, if it worked until now, will continue to work. (And the other way around) >> >> However, if virtio 1.0 is supported by both QEMU and vhost-user, >> virtio 1.0 has different endianess requirements than prev virtio, > > OK so that's an acceptable workaround: big endian platforms must use > virtio 0.9. > > In order to have a big endian support of virtio 1.0, is it sufficient to > convert data? Is it something we want regarding performance?
I think that converting the data should be enough, however Michael can give a more knowledgeable answer, we'll have to wait a few days for him as he is not available this week. Thanks, Marcel >
