[PATCH 2/3] virtio: mmio: fix signature checking for BE guests

[Rusty Russell]

Rusty Russell <rusty at rustcorp.com.au> writes:
> Pawel Moll <pawel.moll at arm.com> writes:
>> On Fri, 2013-10-11 at 15:36 +0100, Marc Zyngier wrote:
>>> As virtio-mmio config registers are specified to be little-endian,
>>> using readl() to read the magic value and then memcmp() to check it
>>> fails on BE (as readl() has an implicit swab).
>>> 
>>> Fix it by encoding the magic value as an integer instead of a string.
>>> 
>>> Cc: Rusty Russell <rusty at rustcorp.com.au>
>>> Cc: Michael S. Tsirkin <mst at redhat.com>
>>> Cc: Pawel Moll <pawel.moll at arm.com>
>>> Signed-off-by: Marc Zyngier <marc.zyngier at arm.com>
>>> ---
>>>  drivers/virtio/virtio_mmio.c | 2 +-
>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>> 
>>> diff --git a/drivers/virtio/virtio_mmio.c b/drivers/virtio/virtio_mmio.c
>>> index 1ba0d68..57f24fd 100644
>>> --- a/drivers/virtio/virtio_mmio.c
>>> +++ b/drivers/virtio/virtio_mmio.c
>>> @@ -470,7 +470,7 @@ static int virtio_mmio_probe(struct platform_device *pdev)
>>>  
>>>  	/* Check magic value */
>>>  	magic = readl(vm_dev->base + VIRTIO_MMIO_MAGIC_VALUE);
>>> -	if (memcmp(&magic, "virt", 4) != 0) {
>>> +	if (magic != ('v' | 'i' << 8 | 'r' << 16 | 't' << 24)) {
>>>  		dev_warn(&pdev->dev, "Wrong magic value 0x%08lx!\n", magic);
>>>  		return -ENODEV;
>>>  	}
>>
>> The new spec will clarify this:
>>
>> * 0x000 | R | MagicValue
>>   Magic value. Must be 0x74726976 (a Little Endian equivalent
>>   of a "virt" string).
>>
>> ... but I like the 'v'i'r't' characters still being there :-)
>>
>> Acked-by: Pawel Moll <pawel.moll at arm.com>
>
> Applied, thanks.

OK, I *said* applied, but left it in my pending queue.

Pawel, do you want to open this can of worms?  If so, I'l merge this
now.

Cheers,
Rusty.