Re: [Qemu-devel] [PATCH v7 2/7] tests: Add virtio device initialization

["Marc Marí" <marc.mari.barcelo@gmail.com>]

El Fri, 29 Aug 2014 16:49:36 +0100
Stefan Hajnoczi <stefanha@gmail.com> escribió:
> On Fri, Aug 29, 2014 at 12:40:43PM +0200, Marc Marí wrote:
> > +static uint64_t qvirtio_pci_config_readq(QVirtioDevice *d, void
> > *addr) +{
> > +    QVirtioPCIDevice *dev = (QVirtioPCIDevice *)d;
> > +    int i;
> > +    union {
> > +        uint8_t bytes[8];
> > +        uint64_t u64;
> > +    } quad;
> > +
> > +    if (qtest_big_endian()) {
> > +        for (i = 0; i < 8; ++i) {
> > +            quad.bytes[7-i] = qpci_io_readb(dev->pdev, addr + i);
> > +        }
> > +    } else {
> > +        for (i = 0; i < 8; ++i) {
> > +            quad.bytes[i] = qpci_io_readb(dev->pdev, addr + i);
> > +        }
> > +    }
> > +
> > +    return quad.u64;
> > +}
> 
> This assumes that the host is little-endian.  The host could be
> big-endian!
> 
> qvirtio_pci_config_readq() needs to return a 64-bit value in host CPU
> endianness.  So the logic should be:
> 
> for (i = 0; i < 8; ++i) {
>     quad.bytes[i] = qpci_io_readb(dev->pdev, addr + i);
> }
> if (qtest_big_endian() != qtest_host_endian()) {
>     quad.u64 = bswap64(quad.u64);
> }
> return quard.u64;
> 
> Stefan

Sorry, I messed up. These are the things that one is not used to think
about, as this is usally taken for granted because the OS handles it.

Marc