[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 0/5] target-arm: ARM64: Adding EL1 AARCH32 guest
From: |
Peter Maydell |
Subject: |
Re: [Qemu-devel] [PATCH 0/5] target-arm: ARM64: Adding EL1 AARCH32 guest support |
Date: |
Tue, 20 Jan 2015 10:26:05 +0000 |
On 20 January 2015 at 10:21, Sergey Fedorov <address@hidden> wrote:
> aarch64-softmmu/qemu-system-aarch64 -M virt -cpu cortex-a57,-aarch64 ...
> It seems a little confusing for me to specify '-aarch64' when forcing
> AArch32 execution state. Why don't just specify 'aarch32' in command line
> instead of '-aarch64' construction?
This is one of the things we discussed during design of this patchset,
and I agree the command line semantics are a bit odd. Essentially, they
make sense from the PoV of the CPU object, because they're saying
"I want a 32-bit only CPU", ie "I do not want the 64 bit feature this
CPU defaults to". Specifying '+aarch32' wouldn't do anything, because
the default Cortex-A57 already has 32-bit support. But from the PoV of
the user, it's a bit odd. The other possible approach to this would be
to have a machine model parameter for "force 32 bit boot", and have this
turn off the CPU 64 bit feature.
Other proposals welcome; I don't like the UI we end up exposing to
the user here, it's just what falls out of the natural way to model
things at the CPU QOM property level.
-- PMM
- [Qemu-devel] [PATCH 2/5] target-arm: Add feature parsing to virt, (continued)
[Qemu-devel] [PATCH 5/5] target-arm: Adjust kernel load address for Image, Greg Bellows, 2015/01/19
Re: [Qemu-devel] [PATCH 0/5] target-arm: ARM64: Adding EL1 AARCH32 guest support, Sergey Fedorov, 2015/01/20
- Re: [Qemu-devel] [PATCH 0/5] target-arm: ARM64: Adding EL1 AARCH32 guest support,
Peter Maydell <=