Re: [PATCH RESEND v8 00/20] riscv: 'max' CPU, detect user choice in TCG

[Andrew Jones]

On Thu, Aug 24, 2023 at 07:14:20PM -0300, Daniel Henrique Barboza wrote:
> Hi,
> 
> This is a resend of these two patch sets because they no longer apply
> into Alistair's riscv-to-apply.next:
> 
> [PATCH v8 00/12] riscv: add 'max' CPU, deprecate 'any'
> 20230815223741.433763-1-dbarboza@ventanamicro.com/">https://lore.kernel.org/qemu-riscv/20230815223741.433763-1-dbarboza@ventanamicro.com/
> 
> [PATCH v3 0/8] riscv: detecting user choice in TCG extensions
> 20230815224733.434682-1-dbarboza@ventanamicro.com/">https://lore.kernel.org/qemu-riscv/20230815224733.434682-1-dbarboza@ventanamicro.com/
> 
> 
> They're being sent in a single package for convenience. No other changes
> made from their old versions aside from the rebase.
> 
> Patches missing acks: 4,7,8
> 
> Changes from v7:
> - patch 7:
>   - add riscv_cpu_add_qdev_prop_array() function instead of a macro
> - patch 8:
>   - add riscv_cpu_add_kvm_unavail_prop_array() function instead of a
>     macro
> - v7 link: 
> 20230815201559.398643-1-dbarboza@ventanamicro.com/">https://lore.kernel.org/qemu-riscv/20230815201559.398643-1-dbarboza@ventanamicro.com/
>

Tracking whether or not the user enabled/disabled an extension is good. I
see we're still not tracking for the MISA extensions, though. If tracking
isn't necessary for them, then a comment above
riscv_cpu_add_misa_properties()
explaining why not would be good. If tracking may be necessary, then
making them consistent with the multi extensions by using the hash would
be good.

Thanks,
drew