[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PULL 1/7] exec: Limit translation limiting in address_
From: |
Peter Maydell |
Subject: |
Re: [Qemu-devel] [PULL 1/7] exec: Limit translation limiting in address_space_translate to xen |
Date: |
Wed, 7 May 2014 16:32:18 +0100 |
On 7 May 2014 16:22, Paolo Bonzini <address@hidden> wrote:
> Fixing Xen to not truncate is not possible because of the Xen mapcache,
> unless of course QEMU is changed to avoid the mapcache completely on 64-bit
> hosts. I'm not sure if that makes sense from the Xen point of view.
>
> Regarding fixing callers, a known one is virtio-scsi which is a bug and on
> my todo list. But another is VFIO, which cannot accept truncation if the
> IOMMU page size is greater than TARGET_PAGE_SIZE.
The API can't simultaneously allow the implementation to truncate
and guarantee to the caller that we don't truncate, so one of these
has to change, surely?
Otherwise we would need to provide some sort of flag for
truncation-unacceptable so that incompatible combinations
fail nicely rather than silently doing weird stuff, I guess.
thanks
-- PMM
[Qemu-devel] [PULL 3/7] xen: move Xen PV machine files to hw/xenpv, Stefano Stabellini, 2014/05/07
[Qemu-devel] [PULL 4/7] xen: move Xen HVM files under hw/i386/xen, Stefano Stabellini, 2014/05/07
[Qemu-devel] [PULL 7/7] xen_disk: add discard support, Stefano Stabellini, 2014/05/07
[Qemu-devel] [PULL 5/7] xen: factor out common functions, Stefano Stabellini, 2014/05/07
Re: [Qemu-devel] [PULL 0/7] xen-140507, Peter Maydell, 2014/05/07