[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RFC PATCH 1/3] target/arm: Implement FEAT_NMI to support Non-maskab
From: |
Jinjie Ruan |
Subject: |
Re: [RFC PATCH 1/3] target/arm: Implement FEAT_NMI to support Non-maskable Interrupt |
Date: |
Tue, 20 Feb 2024 20:49:51 +0800 |
User-agent: |
Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:102.0) Gecko/20100101 Thunderbird/102.2.0 |
On 2024/2/20 20:31, Peter Maydell wrote:
> On Tue, 20 Feb 2024 at 12:19, Jinjie Ruan <ruanjinjie@huawei.com> wrote:
>>
>> Enable Non-maskable Interrupt feature.
>>
>> Enable HCRX register feature to support TALLINT read/write.
>>
>> Add support for enable/disable NMI at qemu startup as below:
>>
>> qemu-system-aarch64 -cpu cortex-a53/a57/a72/a76,nmi=[on/off]
>>
>> Add support for allint read/write as follow:
>>
>> mrs <xt>, ALLINT // read allint
>> msr ALLINT, <xt> // write allint with imm
>> msr ALLINT, #<imm> // write allint with 1 or 0
>
> Can I ask you to break this patchset down into smaller
> coherent pieces, please? When you write a commit message
> that has this sort of "list of four different things the
> patch does" structure, that's a sign that really it ought to
> be multiple different patches that do one thing each.
Thank you very much! I'll break up the patches so that each one does
only one thing.
>
> Do we really need the command line option? Mostly we
> don't add that for new CPU features, unless there's a
> strong reason why users might need to turn it off: instead
> we just implement it if the CPU type and/or the board has
> the feature.
Right! Thank you. I'll remove the command line option and just implement
it if the CPU has the feature.
>
> thanks
> -- PMM