Hi Peter,
Thanks for the reply, yes i am interested in (2). i will follow up on the
suggested mailing list.
Just in case if you know, does the gicv3 support available in QEMU, or
there is a plan.
Regards,
Tirumaqlesh.
On 01-Mar-2014, at 5:58 am, Peter Maydell <peter.mayd...@linaro.org> wrote:
> On 28 February 2014 07:08, Chalamarla, Tirumalesh
> <tirumalesh.chalama...@caviumnetworks.com> wrote:
>> Is there any one, trying out cross compiling and running qemu
>> on aarch64 host. if so is there a development branch where this wrok is
>> progressing.
>>
>> Some one could please let me know the plan/time frame, for
>> qemu on arm64 hosts running arm64 guests.
>
> It's not entirely clear which of the various possible QEMU aarch64
> setups you're interested in. Summary:
>
> (1) using QEMU on aarch64 hosts to emulate other CPU architectures
> (eg x86, MIPS): this went into QEMU about six months ago and was
> in the last release of QEMU (1.7)
>
> (2) using QEMU on aarch64 hosts as the userspace component
> of a VM using KVM kernel support to run an aarch64 guest: this
> should work with current QEMU, though some functionality (for
> instance, migration) is not yet implemented
>
> (3) using QEMU to emulate individual Linux AArch64 binaries, running
> on any host (typically x86): this works in current upstream master, but
> some instructions (parts of SIMD) are not yet implemented. I hope
> we'll get the SIMD coverage completed within the next few weeks,
> in time to put it into QEMU 2.0.
>
> (4) using QEMU to emulate an entire AArch64 system that can
> boot a guest kernel, typically running on an x86 host: we're
> working on this right now; we have work-in-progress code which
> will boot a kernel and are working on cleaning it up to upstream
> quality. I expect we'll have this done within a month or two, but
> it won't make it into the QEMU 2.0 release (slightly too late).
>
> I'm guessing you're interested in (2) or maybe (1). For (2),
> kvm...@lists.cs.columbia.edu is a good list to follow to
> monitor what's currently going on.
>
> thanks
> -- PMM
