Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-06 Thread Janosch Frank
On 3/6/20 12:48 PM, Christian Borntraeger wrote: > > > On 04.03.20 12:42, Janosch Frank wrote: > [...] >> diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c >> index 3dd396e870..69b1cc5dfc 100644 >> --- a/target/s390x/cpu.c >> +++ b/target/s390x/cpu.c >> @@ -37,6 +37,8 @@ >> #include

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-06 Thread Christian Borntraeger
On 04.03.20 12:42, Janosch Frank wrote: [...] > diff --git a/target/s390x/cpu.c b/target/s390x/cpu.c > index 3dd396e870..69b1cc5dfc 100644 > --- a/target/s390x/cpu.c > +++ b/target/s390x/cpu.c > @@ -37,6 +37,8 @@ > #include "sysemu/hw_accel.h" > #include "hw/qdev-properties.h" > #ifndef

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-05 Thread Janosch Frank
On 3/5/20 3:23 PM, David Hildenbrand wrote: > On 05.03.20 15:20, Janosch Frank wrote: >> On 3/5/20 3:15 PM, David Hildenbrand wrote: >> >> +static void s390_machine_unprotect(S390CcwMachineState *ms) >> +{ >> +CPUState *t; >> + >> +if (!ms->pv) >> +

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-05 Thread David Hildenbrand
On 05.03.20 15:20, Janosch Frank wrote: > On 3/5/20 3:15 PM, David Hildenbrand wrote: > > +static void s390_machine_unprotect(S390CcwMachineState *ms) > +{ > +CPUState *t; > + > +if (!ms->pv) > +return; How can this ever happen?

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-05 Thread Janosch Frank
On 3/5/20 3:15 PM, David Hildenbrand wrote: +static void s390_machine_unprotect(S390CcwMachineState *ms) +{ +CPUState *t; + +if (!ms->pv) +return; >>> >>> How can this ever happen? g_assert(ms->pv) ? >> >> Currently not, that's only used in the

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-05 Thread Janosch Frank
On 3/5/20 2:52 PM, David Hildenbrand wrote: > On 04.03.20 12:42, Janosch Frank wrote: >> When a guest has saved a ipib of type 5 and calls diagnose308 with >> subcode 10, we have to setup the protected processing environment via >> Ultravisor calls. The calls are done by KVM and are exposed via an

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-05 Thread David Hildenbrand
>>> >>> +static void s390_machine_unprotect(S390CcwMachineState *ms) >>> +{ >>> +CPUState *t; >>> + >>> +if (!ms->pv) >>> +return; >> >> How can this ever happen? g_assert(ms->pv) ? > > Currently not, that's only used in the follow up patches with the ballon > and migration

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-05 Thread Janosch Frank
On 3/5/20 2:51 PM, David Hildenbrand wrote: > On 04.03.20 12:42, Janosch Frank wrote: >> When a guest has saved a ipib of type 5 and calls diagnose308 with > > s/a/an/ > >> subcode 10, we have to setup the protected processing environment via >> Ultravisor calls. The calls are done by KVM and

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-05 Thread David Hildenbrand
On 04.03.20 12:42, Janosch Frank wrote: > When a guest has saved a ipib of type 5 and calls diagnose308 with s/a/an/ > subcode 10, we have to setup the protected processing environment via > Ultravisor calls. The calls are done by KVM and are exposed via an > API. > > The following steps are

Re: [PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-05 Thread David Hildenbrand
On 04.03.20 12:42, Janosch Frank wrote: > When a guest has saved a ipib of type 5 and calls diagnose308 with > subcode 10, we have to setup the protected processing environment via > Ultravisor calls. The calls are done by KVM and are exposed via an > API. > > The following steps are necessary: >

[PATCH v6 03/18] s390x: protvirt: Support unpack facility

2020-03-04 Thread Janosch Frank
When a guest has saved a ipib of type 5 and calls diagnose308 with subcode 10, we have to setup the protected processing environment via Ultravisor calls. The calls are done by KVM and are exposed via an API. The following steps are necessary: 1. Enable protected mode for the VM (register it and