Actually, it is my fault. I had the same issue when I tested the initial patchset (v1) but for some reason the fix has not been merged into the patchset I sent to the EDK2 mailing-list.
This is the way I have validated the VirtIo changes: 1) Create two disk images ------- mkfs.vfat -C -n IMAGE /tmp/block.dat 131072 mkdir /tmp/fs sudo mount -o rw,loop=/dev/loop0,uid=`whoami`,gid=`whoami` /tmp/block.dat /tmp/fs cp $EDK2_ROOT/EdkShellBinPkg/FullShell/Arm/Shell_Full.efi /tmp/fs/ touch /tmp/fs/hello.txt sudo umount /tmp/fs ------- 2) qEmu version commit c1990468d5a13c8f1d9111fcca0a78a6adad062b Author: Michael Roth <mdr...@linux.vnet.ibm.com> Date: Wed May 22 11:32:51 2013 -0500 3) Build OvmfPkg: ./OvmfPkg/build.sh 4) Run qEmu: PATH=/work/qemu/x86_64-softmmu:$PATH OvmfPkg/build.sh -a X64 qemu -serial file:serial.log -debugcon stdio -global isa-debugcon.iobase=0x402 -drive file=/tmp/block.dat,if=virtio,id=blk -drive file=/tmp/scsi.dat,if=none,id=scsi -device virtio-scsi-pci,id=fred -device scsi-hd,drive=scsi -net nic,model=virtio -net user 5) Go to the Shell - Ensure the VirtIo BlockIo and SCSI file systems are readable and writable - Ensure an IP can be retrieved for the VirtIo Network device Olivier > -----Original Message----- > From: Laszlo Ersek [mailto:ler...@redhat.com] > Sent: 20 September 2013 17:03 > To: Paolo Bonzini > Cc: edk2-devel@lists.sourceforge.net; Olivier Martin > Subject: Re: [edk2] [PATCH v2 04/10] OvmfPkg/VirtioPciDeviceDxe: > Implement VIRTIO_DEVICE_PROTOCOL for VirtIo Devices over PCI > > On 09/20/13 17:42, Paolo Bonzini wrote: > > Il 19/09/2013 22:30, Laszlo Ersek ha scritto: > >> In a nutshell, VirtioPciSetDeviceStatus() breaks the interface > contract > >> of VIRTIO_DEVICE_PROTOCOL.WriteDevice(). The former targets a common > >> virtio field, but the latter provides access to device-specific > fields. > >> Consequently, the virtio device is never brought up, and the guest > code > >> in VirtioFlush() waits for the hypervisor's reply in vain. > >> > >> I don't yet know enough about the series to suggest a fix (or to fix > it > >> myself just so I can continue the testing and the review). I'd like > to > >> get good test results before reviewing in more detail. > > > > Maybe VirtioPciSetDeviceStatus and the other accesses to the common > > virtio fields should use PciIo directly? > > Yes. Or maybe continue using the current PciIo wrapper, but expose the > offset base as a parameter (or extract a common helper function etc.) > > Thanks, > Laszlo ------------------------------------------------------------------------------ LIMITED TIME SALE - Full Year of Microsoft Training For Just $49.99! 1,500+ hours of tutorials including VisualStudio 2012, Windows 8, SharePoint 2013, SQL 2012, MVC 4, more. BEST VALUE: New Multi-Library Power Pack includes Mobile, Cloud, Java, and UX Design. Lowest price ever! Ends 9/20/13. http://pubads.g.doubleclick.net/gampad/clk?id=58041151&iu=/4140/ostg.clktrk _______________________________________________ edk2-devel mailing list edk2-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/edk2-devel
