Re: [Xen-devel] [PATCH 32/34] x86: expose CONFIG_HVM

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 21.08.18 at 14:49, <andrew.cooper3@xxxxxxxxxx> wrote:
> On 21/08/2018 12:48, Jan Beulich wrote:
>>>>> On 17.08.18 at 17:12, <wei.liu2@xxxxxxxxxx> wrote:
>>> --- a/xen/arch/x86/Kconfig
>>> +++ b/xen/arch/x86/Kconfig
>>> @@ -60,6 +60,12 @@ config PV_LINEAR_PT
>>>  
>>>  config HVM
>>>     def_bool y
>>> +   prompt "HVM / PVH support"
>>> +       ---help---
>>> +         Interfaces to support HVM and PVH guests.
>>> +
>>> +         If unsure, say Y.
>> As said before, this wants to gain a dependency on !PV_SHIM_EXCLUSIVE.
> 
> No - it should not.
> 
> Dependencies force options to be hidden in Kconfig, which lead to very
> poor use experience for people trying to find an option they expect to
> be available.
> 
> There no functional dependency between HVM and PV_SHIM_EXCLUSIVE, so
> there should be no dependency in Kconfig.  PV_SHIM_EXCLUSIVE should
> probably recommend in its help text to disable HVM, but that is the most
> which Kconfig should influence.

I disagree; at the very least the default of the option here should
not be Yes when PV_SHIM_EXCLUSIVE is enabled. But I think a
depends is The Right Thing To Do (tm) here; user experience when
using the various frontends is only one aspect, users getting their
selections wrong because they've been given too much control is
another. What business would a user wanting to configure the shim
have finding the HVM option?

Jan



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel