RE: [PATCH v3 2/7] Drivers: hv: Enable VTL mode for arm64

[Roman Kisel <romank@linux.microsoft.com>]

> > 
> > On 8/4/2024 9:05 PM, Saurabh Singh Sengar wrote:
> > > On Mon, Aug 05, 2024 at 03:01:58AM +0000, Michael Kelley wrote:
> > >> From: Roman Kisel <romank@linux.microsoft.com> Sent: Friday, July 26, 2024 3:59
> > PM
> > >>>
> > >>> Kconfig dependencies for arm64 guests on Hyper-V require that be ACPI enabled,
> > >>> and limit VTL mode to x86/x64. To enable VTL mode on arm64 as well, update the
> > >>> dependencies. Since VTL mode requires DeviceTree instead of ACPI, don't require
> > >>> arm64 guests on Hyper-V to have ACPI.
> > >>>
> > >>> Signed-off-by: Roman Kisel <romank@linux.microsoft.com>
> > >>> ---
> > >>>   drivers/hv/Kconfig | 6 +++---
> > >>>   1 file changed, 3 insertions(+), 3 deletions(-)
> > >>>
> > >>> diff --git a/drivers/hv/Kconfig b/drivers/hv/Kconfig
> > >>> index 862c47b191af..a5cd1365e248 100644
> > >>> --- a/drivers/hv/Kconfig
> > >>> +++ b/drivers/hv/Kconfig
> > >>> @@ -5,7 +5,7 @@ menu "Microsoft Hyper-V guest support"
> > >>>   config HYPERV
> > >>>   	tristate "Microsoft Hyper-V client drivers"
> > >>>   	depends on (X86 && X86_LOCAL_APIC && HYPERVISOR_GUEST) \
> > >>> -		|| (ACPI && ARM64 && !CPU_BIG_ENDIAN)
> > >>> +		|| (ARM64 && !CPU_BIG_ENDIAN)
> > >>>   	select PARAVIRT
> > >>>   	select X86_HV_CALLBACK_VECTOR if X86
> > >>>   	select OF_EARLY_FLATTREE if OF
> > >>> @@ -15,7 +15,7 @@ config HYPERV
> > >>>
> > >>>   config HYPERV_VTL_MODE
> > >>>   	bool "Enable Linux to boot in VTL context"
> > >>> -	depends on X86_64 && HYPERV
> > >>> +	depends on HYPERV
> > >>>   	depends on SMP
> > >>>   	default n
> > >>>   	help
> > >>> @@ -31,7 +31,7 @@ config HYPERV_VTL_MODE
> > >>>
> > >>>   	  Select this option to build a Linux kernel to run at a VTL other than
> > >>>   	  the normal VTL0, which currently is only VTL2.  This option
> > >>> -	  initializes the x86 platform for VTL2, and adds the ability to boot
> > >>> +	  initializes the kernel to run in VTL2, and adds the ability to boot
> > >>>   	  secondary CPUs directly into 64-bit context as required for VTLs other
> > >>>   	  than 0.  A kernel built with this option must run at VTL2, and will
> > >>>   	  not run as a normal guest.
> > >>> --
> > >>> 2.34.1
> > >>>
> > >>
> > >> In v2 of this patch, I suggested [1] making a couple additional minor changes
> > >> so that kernels built *without* HYPER_VTL_MODE would still require
> > >> ACPI.  Did that suggestion not work out?  If that's the case, I'm curious
> > >> about what goes wrong.
> > >
> > > Hi Michael/Roman,
> > > I was considering making HYPERV_VTL_MODE depend on CONFIG_OF. That should
> > address
> > > above concern as well. Do you see any potential issue with it.
> > >
> > Michael,
> > 
> > I ran into a pretty gnarly recursive dependencies which in all fairness
> > might stem from not being fluent enough in the Kconfig language. Any
> > help of how to approach implementing your idea would be greatly appreciated!
> > 
> 
> This is what I had in mind:
> 
> --- a/drivers/hv/Kconfig
> +++ b/drivers/hv/Kconfig
> @@ -5,7 +5,8 @@ menu "Microsoft Hyper-V guest support"
 > config HYPERV
        > tristate "Microsoft Hyper-V client drivers"
        > depends on (X86 && X86_LOCAL_APIC && HYPERVISOR_GUEST) \
> -               || (ACPI && ARM64 && !CPU_BIG_ENDIAN)
> +               || (ARM64 && !CPU_BIG_ENDIAN)
> +       depends on (ACPI || HYPERV_VTL_MODE)
        > select PARAVIRT
        > select X86_HV_CALLBACK_VECTOR if X86
        > select OF_EARLY_FLATTREE if OF
> @@ -15,7 +16,7 @@ config HYPERV
> 
 > config HYPERV_VTL_MODE
        > bool "Enable Linux to boot in VTL context"
> -       depends on X86_64 && HYPERV
> +       depends on X86_64
        > depends on SMP
        > default n
        > help
> 
> HYPERV_VTL_MODE can now be selected independently of HYPERV.
> The existing code should be such that even if someone is building a
> random config and gets HYPERV_VTL_MODE without HYPERV, the
> kernel will build and run in a non-Hyper-V environment and isn't
> broken somehow.
> 
> For HYPERV to be selected, either ACPI must already be selected, or
> HYPERV_VTL_MODE must already be selected. So "normal" kernels are
> still enforced to require ACPI. But if building with HYPERV_VTL_MODE,
> then ACPI is optional.
> 
Thanks a ton! Let me try this out for the (arch; VTL) build matrix :)

> Saurabh's idea of adding "depends on OF" to HYPERV_VTL_MODE
> should also work with these changes.
> 
> I haven't fully tested the above with all the relevant combinations
> on both x86 and arm64, but I think the logic makes sense.