Re: [kvm-unit-tests PATCH 7/8] s390x: snippets: Add PV support

[Claudio Imbrenda <imbrenda@linux.ibm.com>]

On Thu, 13 Jan 2022 14:10:01 +0100
Janosch Frank <frankja@linux.ibm.com> wrote:

> On 12/3/21 10:29, Janosch Frank wrote:
> > On 11/26/21 14:28, Janosch Frank wrote:  
> >> On 11/23/21 12:22, Claudio Imbrenda wrote:  
> >>> On Tue, 23 Nov 2021 10:39:55 +0000
> >>> Janosch Frank <frankja@linux.ibm.com> wrote:
> >>>  
> [...]
> >>> are they supposed to have different addresses?
> >>> the C files start at 0x4000, while the asm ones at 0  
> >>
> >> That's a mistake I'll need to fix.  
> > 
> > On second thought this is correct since it's the starting address of the
> > component and not the PSW entry. The psw entry is the next argument.
> > The C snippets currently have data in the first 4 pages so we can
> > encrypt from offset 0.
> > 
> > The question that remains is: do we need the data at 0x0 - 0x4000?
> > The reset and restart PSWs are not really necessary since we don't start
> > the snippets as a lpar or in simulation where we use these PSWs.
> > The stackptr is just that, a ptr AFAIK so there shouldn't be data on
> > 0x3000 (but I'll look that up anyway).  
> 
> Colleagues have used the C PV snippets over the last few weeks and 
> haven't reported any issues. It's time that we bring this into master 
> since a lot of upcoming tests are currently based on this series.
> 
> @Claudio: Any further comments?

no, let's bring this into master