| Message ID | 1544135058-21380-3-git-send-email-walling@linux.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Guest Support for Diagnose 318 | expand |
On Thu, 6 Dec 2018 17:24:17 -0500 Collin Walling <walling@linux.ibm.com> wrote: > Diagnose 318 is a new z14.2 CPU feature. Since we are able to emulate > it entirely via KVM, we can add guest support for earlier models. A > new CPU feature for diagnose 318 (shortened to diag318) will be made > available to guests starting with the zEC12-full CPU model. > > The z14.2 adds a new read SCP info byte (let's call it byte 134) to > detect the availability of diag318. Because of this, we have room for > one less VCPU and thus limit the max VPUs supported in a configuration > to 247 (down from 248). > > Signed-off-by: Collin Walling <walling@linux.ibm.com>. > --- > hw/s390x/sclp.c | 2 ++ > include/hw/s390x/sclp.h | 2 ++ > target/s390x/cpu.h | 2 +- > target/s390x/cpu_features.c | 3 +++ > target/s390x/cpu_features.h | 1 + > target/s390x/cpu_features_def.h | 3 +++ > target/s390x/gen-features.c | 1 + > target/s390x/kvm.c | 1 + > 8 files changed, 14 insertions(+), 1 deletion(-) > > diff --git a/target/s390x/cpu.h b/target/s390x/cpu.h > index 8c2320e..594b4a4 100644 > --- a/target/s390x/cpu.h > +++ b/target/s390x/cpu.h > @@ -52,7 +52,7 @@ > > #define MMU_USER_IDX 0 > > -#define S390_MAX_CPUS 248 > +#define S390_MAX_CPUS 247 Isn't that already problematic if you try to migrate from an older QEMU with all possible vcpus defined? IOW, don't you really need a way that older machines can still run with one more vcpu? > > typedef struct PSW { > uint64_t mask;
On 12/7/18 7:08 AM, Cornelia Huck wrote: > On Thu, 6 Dec 2018 17:24:17 -0500 > Collin Walling <walling@linux.ibm.com> wrote: > >> Diagnose 318 is a new z14.2 CPU feature. Since we are able to emulate >> it entirely via KVM, we can add guest support for earlier models. A >> new CPU feature for diagnose 318 (shortened to diag318) will be made >> available to guests starting with the zEC12-full CPU model. >> >> The z14.2 adds a new read SCP info byte (let's call it byte 134) to >> detect the availability of diag318. Because of this, we have room for >> one less VCPU and thus limit the max VPUs supported in a configuration >> to 247 (down from 248). >> >> Signed-off-by: Collin Walling <walling@linux.ibm.com>. >> --- >> hw/s390x/sclp.c | 2 ++ >> include/hw/s390x/sclp.h | 2 ++ >> target/s390x/cpu.h | 2 +- >> target/s390x/cpu_features.c | 3 +++ >> target/s390x/cpu_features.h | 1 + >> target/s390x/cpu_features_def.h | 3 +++ >> target/s390x/gen-features.c | 1 + >> target/s390x/kvm.c | 1 + >> 8 files changed, 14 insertions(+), 1 deletion(-) >> > >> diff --git a/target/s390x/cpu.h b/target/s390x/cpu.h >> index 8c2320e..594b4a4 100644 >> --- a/target/s390x/cpu.h >> +++ b/target/s390x/cpu.h >> @@ -52,7 +52,7 @@ >> >> #define MMU_USER_IDX 0 >> >> -#define S390_MAX_CPUS 248 >> +#define S390_MAX_CPUS 247 > > Isn't that already problematic if you try to migrate from an older QEMU > with all possible vcpus defined? IOW, don't you really need a way that > older machines can still run with one more vcpu? > Good call. I'll run some tests on this and see what happens. I'll report here on those results. >> >> typedef struct PSW { >> uint64_t mask; >
On 12/11/18 11:47 AM, Collin Walling wrote: > On 12/7/18 7:08 AM, Cornelia Huck wrote: >> On Thu, 6 Dec 2018 17:24:17 -0500 >> Collin Walling <walling@linux.ibm.com> wrote: >> >>> Diagnose 318 is a new z14.2 CPU feature. Since we are able to emulate >>> it entirely via KVM, we can add guest support for earlier models. A >>> new CPU feature for diagnose 318 (shortened to diag318) will be made >>> available to guests starting with the zEC12-full CPU model. >>> >>> The z14.2 adds a new read SCP info byte (let's call it byte 134) to >>> detect the availability of diag318. Because of this, we have room for >>> one less VCPU and thus limit the max VPUs supported in a configuration >>> to 247 (down from 248). >>> >>> Signed-off-by: Collin Walling <walling@linux.ibm.com>. >>> --- >>> hw/s390x/sclp.c | 2 ++ >>> include/hw/s390x/sclp.h | 2 ++ >>> target/s390x/cpu.h | 2 +- >>> target/s390x/cpu_features.c | 3 +++ >>> target/s390x/cpu_features.h | 1 + >>> target/s390x/cpu_features_def.h | 3 +++ >>> target/s390x/gen-features.c | 1 + >>> target/s390x/kvm.c | 1 + >>> 8 files changed, 14 insertions(+), 1 deletion(-) >>> >> >>> diff --git a/target/s390x/cpu.h b/target/s390x/cpu.h >>> index 8c2320e..594b4a4 100644 >>> --- a/target/s390x/cpu.h >>> +++ b/target/s390x/cpu.h >>> @@ -52,7 +52,7 @@ >>> >>> #define MMU_USER_IDX 0 >>> >>> -#define S390_MAX_CPUS 248 >>> +#define S390_MAX_CPUS 247 >> >> Isn't that already problematic if you try to migrate from an older QEMU >> with all possible vcpus defined? IOW, don't you really need a way that >> older machines can still run with one more vcpu? >> > > Good call. I'll run some tests on this and see what happens. I'll report > here on those results. > Migrating to a machine that supports less vCPUs will report error: unsupported configuration: Maximum CPUs greater than specified machine type limit I revisited the code to see if there's a way to dynamically set the max vcpu count based on the read scp info size, but it gets really tricky and code looks very complicated. (Having a packed struct contain the CPU entries whose maximum is determined by hardware limitations makes things difficult -- but who said s390 is easy? :) ) In reality, do we often have guests running with 248 or even 247 vcpus? If so, I imagine the performance isn't too significant? >>> >>> typedef struct PSW { >>> uint64_t mask; >> > >
On 11.12.18 22:12, Collin Walling wrote: > On 12/11/18 11:47 AM, Collin Walling wrote: >> On 12/7/18 7:08 AM, Cornelia Huck wrote: >>> On Thu, 6 Dec 2018 17:24:17 -0500 >>> Collin Walling <walling@linux.ibm.com> wrote: >>> >>>> Diagnose 318 is a new z14.2 CPU feature. Since we are able to emulate >>>> it entirely via KVM, we can add guest support for earlier models. A >>>> new CPU feature for diagnose 318 (shortened to diag318) will be made >>>> available to guests starting with the zEC12-full CPU model. >>>> >>>> The z14.2 adds a new read SCP info byte (let's call it byte 134) to >>>> detect the availability of diag318. Because of this, we have room for >>>> one less VCPU and thus limit the max VPUs supported in a configuration >>>> to 247 (down from 248). >>>> >>>> Signed-off-by: Collin Walling <walling@linux.ibm.com>. >>>> --- >>>> hw/s390x/sclp.c | 2 ++ >>>> include/hw/s390x/sclp.h | 2 ++ >>>> target/s390x/cpu.h | 2 +- >>>> target/s390x/cpu_features.c | 3 +++ >>>> target/s390x/cpu_features.h | 1 + >>>> target/s390x/cpu_features_def.h | 3 +++ >>>> target/s390x/gen-features.c | 1 + >>>> target/s390x/kvm.c | 1 + >>>> 8 files changed, 14 insertions(+), 1 deletion(-) >>>> >>> >>>> diff --git a/target/s390x/cpu.h b/target/s390x/cpu.h >>>> index 8c2320e..594b4a4 100644 >>>> --- a/target/s390x/cpu.h >>>> +++ b/target/s390x/cpu.h >>>> @@ -52,7 +52,7 @@ >>>> >>>> #define MMU_USER_IDX 0 >>>> >>>> -#define S390_MAX_CPUS 248 >>>> +#define S390_MAX_CPUS 247 >>> >>> Isn't that already problematic if you try to migrate from an older QEMU >>> with all possible vcpus defined? IOW, don't you really need a way that >>> older machines can still run with one more vcpu? >>> >> >> Good call. I'll run some tests on this and see what happens. I'll report >> here on those results. >> > > Migrating to a machine that supports less vCPUs will report > > error: unsupported configuration: Maximum CPUs greater than specified machine type limit > > I revisited the code to see if there's a way to dynamically set the max vcpu count based > on the read scp info size, but it gets really tricky and code looks very complicated. > (Having a packed struct contain the CPU entries whose maximum is determined by hardware > limitations makes things difficult -- but who said s390 is easy? :) ) > > In reality, do we often have guests running with 248 or even 247 vcpus? If so, I imagine > the performance isn't too significant? Gluing CPU feature availability to machines is plain ugly. This sounds like going back to pre-cpu model times ;) There are two alternatives: a) Don't model it as a CPU feature in QEMU. Glue it completely to the QEMU machine. This goes hand-in-hand with the proposal I made in the KVM thread, that diag318 is to be handled completely in QEMU, always. The KVM setting part is optional (if KVM + HW support it). Then we can have two different max_cpus/ReadInfo layouts based on the machine type. No need to worry about QEMU cpu features. Once we have other SCLP features (eventually requiring KVM/HW support) announced in the same feature block, things might get more involved, but I guess we could handle it somehow. b) Glue the ReadInfo layout to the CPU feature, we would have to default-disable the CPU feature for legacy machines. And bail out if more CPUs are used when the feature is enabled. Hairy. I guess a) would be the best thing to do. After all this really does not sound like a CPU feature but more like a machine feature. But there is usually a fine line between them.
On Wed, 12 Dec 2018 12:20:08 +0100 David Hildenbrand <david@redhat.com> wrote: > On 11.12.18 22:12, Collin Walling wrote: > > On 12/11/18 11:47 AM, Collin Walling wrote: > >> On 12/7/18 7:08 AM, Cornelia Huck wrote: > >>> On Thu, 6 Dec 2018 17:24:17 -0500 > >>> Collin Walling <walling@linux.ibm.com> wrote: > >>> > >>>> Diagnose 318 is a new z14.2 CPU feature. Since we are able to emulate > >>>> it entirely via KVM, we can add guest support for earlier models. A > >>>> new CPU feature for diagnose 318 (shortened to diag318) will be made > >>>> available to guests starting with the zEC12-full CPU model. > >>>> > >>>> The z14.2 adds a new read SCP info byte (let's call it byte 134) to > >>>> detect the availability of diag318. Because of this, we have room for > >>>> one less VCPU and thus limit the max VPUs supported in a configuration > >>>> to 247 (down from 248). > >>>> > >>>> Signed-off-by: Collin Walling <walling@linux.ibm.com>. > >>>> --- > >>>> hw/s390x/sclp.c | 2 ++ > >>>> include/hw/s390x/sclp.h | 2 ++ > >>>> target/s390x/cpu.h | 2 +- > >>>> target/s390x/cpu_features.c | 3 +++ > >>>> target/s390x/cpu_features.h | 1 + > >>>> target/s390x/cpu_features_def.h | 3 +++ > >>>> target/s390x/gen-features.c | 1 + > >>>> target/s390x/kvm.c | 1 + > >>>> 8 files changed, 14 insertions(+), 1 deletion(-) > >>>> > >>> > >>>> diff --git a/target/s390x/cpu.h b/target/s390x/cpu.h > >>>> index 8c2320e..594b4a4 100644 > >>>> --- a/target/s390x/cpu.h > >>>> +++ b/target/s390x/cpu.h > >>>> @@ -52,7 +52,7 @@ > >>>> > >>>> #define MMU_USER_IDX 0 > >>>> > >>>> -#define S390_MAX_CPUS 248 > >>>> +#define S390_MAX_CPUS 247 > >>> > >>> Isn't that already problematic if you try to migrate from an older QEMU > >>> with all possible vcpus defined? IOW, don't you really need a way that > >>> older machines can still run with one more vcpu? > >>> > >> > >> Good call. I'll run some tests on this and see what happens. I'll report > >> here on those results. > >> > > > > Migrating to a machine that supports less vCPUs will report > > > > error: unsupported configuration: Maximum CPUs greater than specified machine type limit > > > > I revisited the code to see if there's a way to dynamically set the max vcpu count based > > on the read scp info size, but it gets really tricky and code looks very complicated. > > (Having a packed struct contain the CPU entries whose maximum is determined by hardware > > limitations makes things difficult -- but who said s390 is easy? :) ) > > > > In reality, do we often have guests running with 248 or even 247 vcpus? If so, I imagine > > the performance isn't too significant? > Gluing CPU feature availability to machines is plain ugly. This sounds > like going back to pre-cpu model times ;) > > There are two alternatives: > > a) Don't model it as a CPU feature in QEMU. Glue it completely to the > QEMU machine. This goes hand-in-hand with the proposal I made in the KVM > thread, that diag318 is to be handled completely in QEMU, always. The > KVM setting part is optional (if KVM + HW support it). > > Then we can have two different max_cpus/ReadInfo layouts based on the > machine type. No need to worry about QEMU cpu features. > > Once we have other SCLP features (eventually requiring KVM/HW support) > announced in the same feature block, things might get more involved, but > I guess we could handle it somehow. Perhaps via a capability to be enabled? > > > b) Glue the ReadInfo layout to the CPU feature, we would have to > default-disable the CPU feature for legacy machines. And bail out if > more CPUs are used when the feature is enabled. Hairy. > > > I guess a) would be the best thing to do. After all this really does not > sound like a CPU feature but more like a machine feature. But there is > usually a fine line between them. a) sounds like the better option to me as well.
On 12/12/18 8:41 AM, Cornelia Huck wrote: > On Wed, 12 Dec 2018 12:20:08 +0100 > David Hildenbrand <david@redhat.com> wrote: > >> On 11.12.18 22:12, Collin Walling wrote: >>> On 12/11/18 11:47 AM, Collin Walling wrote: >>>> On 12/7/18 7:08 AM, Cornelia Huck wrote: >>>>> On Thu, 6 Dec 2018 17:24:17 -0500 >>>>> Collin Walling <walling@linux.ibm.com> wrote: >>>>> >>>>>> Diagnose 318 is a new z14.2 CPU feature. Since we are able to emulate >>>>>> it entirely via KVM, we can add guest support for earlier models. A >>>>>> new CPU feature for diagnose 318 (shortened to diag318) will be made >>>>>> available to guests starting with the zEC12-full CPU model. >>>>>> >>>>>> The z14.2 adds a new read SCP info byte (let's call it byte 134) to >>>>>> detect the availability of diag318. Because of this, we have room for >>>>>> one less VCPU and thus limit the max VPUs supported in a configuration >>>>>> to 247 (down from 248). >>>>>> >>>>>> Signed-off-by: Collin Walling <walling@linux.ibm.com>. >>>>>> --- >>>>>> hw/s390x/sclp.c | 2 ++ >>>>>> include/hw/s390x/sclp.h | 2 ++ >>>>>> target/s390x/cpu.h | 2 +- >>>>>> target/s390x/cpu_features.c | 3 +++ >>>>>> target/s390x/cpu_features.h | 1 + >>>>>> target/s390x/cpu_features_def.h | 3 +++ >>>>>> target/s390x/gen-features.c | 1 + >>>>>> target/s390x/kvm.c | 1 + >>>>>> 8 files changed, 14 insertions(+), 1 deletion(-) >>>>>> >>>>> >>>>>> diff --git a/target/s390x/cpu.h b/target/s390x/cpu.h >>>>>> index 8c2320e..594b4a4 100644 >>>>>> --- a/target/s390x/cpu.h >>>>>> +++ b/target/s390x/cpu.h >>>>>> @@ -52,7 +52,7 @@ >>>>>> >>>>>> #define MMU_USER_IDX 0 >>>>>> >>>>>> -#define S390_MAX_CPUS 248 >>>>>> +#define S390_MAX_CPUS 247 >>>>> >>>>> Isn't that already problematic if you try to migrate from an older QEMU >>>>> with all possible vcpus defined? IOW, don't you really need a way that >>>>> older machines can still run with one more vcpu? >>>>> >>>> >>>> Good call. I'll run some tests on this and see what happens. I'll report >>>> here on those results. >>>> >>> >>> Migrating to a machine that supports less vCPUs will report >>> >>> error: unsupported configuration: Maximum CPUs greater than specified machine type limit >>> >>> I revisited the code to see if there's a way to dynamically set the max vcpu count based >>> on the read scp info size, but it gets really tricky and code looks very complicated. >>> (Having a packed struct contain the CPU entries whose maximum is determined by hardware >>> limitations makes things difficult -- but who said s390 is easy? :) ) >>> >>> In reality, do we often have guests running with 248 or even 247 vcpus? If so, I imagine >>> the performance isn't too significant? >> Gluing CPU feature availability to machines is plain ugly. This sounds >> like going back to pre-cpu model times ;) >> >> There are two alternatives: >> >> a) Don't model it as a CPU feature in QEMU. Glue it completely to the >> QEMU machine. This goes hand-in-hand with the proposal I made in the KVM >> thread, that diag318 is to be handled completely in QEMU, always. The >> KVM setting part is optional (if KVM + HW support it). >> >> Then we can have two different max_cpus/ReadInfo layouts based on the >> machine type. No need to worry about QEMU cpu features. >> >> Once we have other SCLP features (eventually requiring KVM/HW support) >> announced in the same feature block, things might get more involved, but >> I guess we could handle it somehow. > > Perhaps via a capability to be enabled? > >> >> >> b) Glue the ReadInfo layout to the CPU feature, we would have to >> default-disable the CPU feature for legacy machines. And bail out if >> more CPUs are used when the feature is enabled. Hairy. >> >> >> I guess a) would be the best thing to do. After all this really does not >> sound like a CPU feature but more like a machine feature. But there is >> usually a fine line between them. > > a) sounds like the better option to me as well. > I think this makes sense as well. A CPU feat really doesn't make sense if we just want to enable this "always" so-to-speak. I'll get cracking on a rework of this patch series. It'll take me some time. In the mean time, I'll return the favor and take a look at the PCI stuff you guys have posted ;)
On 06.12.2018 23:24, Collin Walling wrote: > Diagnose 318 is a new z14.2 CPU feature. Since we are able to emulate > it entirely via KVM, we can add guest support for earlier models. A > new CPU feature for diagnose 318 (shortened to diag318) will be made > available to guests starting with the zEC12-full CPU model. > > The z14.2 adds a new read SCP info byte (let's call it byte 134) to > detect the availability of diag318. Because of this, we have room for > one less VCPU and thus limit the max VPUs supported in a configuration > to 247 (down from 248). > > Signed-off-by: Collin Walling <walling@linux.ibm.com>. > --- > hw/s390x/sclp.c | 2 ++ > include/hw/s390x/sclp.h | 2 ++ > target/s390x/cpu.h | 2 +- > target/s390x/cpu_features.c | 3 +++ > target/s390x/cpu_features.h | 1 + > target/s390x/cpu_features_def.h | 3 +++ > target/s390x/gen-features.c | 1 + > target/s390x/kvm.c | 1 + > 8 files changed, 14 insertions(+), 1 deletion(-) > > diff --git a/hw/s390x/sclp.c b/hw/s390x/sclp.c > index 4510a80..183c627 100644 > --- a/hw/s390x/sclp.c > +++ b/hw/s390x/sclp.c > @@ -73,6 +73,8 @@ static void read_SCP_info(SCLPDevice *sclp, SCCB *sccb) > read_info->conf_char); > s390_get_feat_block(S390_FEAT_TYPE_SCLP_CONF_CHAR_EXT, > read_info->conf_char_ext); > + /* Read Info byte 134 */ > + s390_get_feat_block(S390_FEAT_TYPE_SCLP_BYTE_134, read_info->byte_134); > > read_info->facilities = cpu_to_be64(SCLP_HAS_CPU_INFO | > SCLP_HAS_IOA_RECONFIG); > diff --git a/include/hw/s390x/sclp.h b/include/hw/s390x/sclp.h > index f9db243..eb12ba2 100644 > --- a/include/hw/s390x/sclp.h > +++ b/include/hw/s390x/sclp.h > @@ -133,6 +133,8 @@ typedef struct ReadInfo { > uint16_t highest_cpu; > uint8_t _reserved5[124 - 122]; /* 122-123 */ > uint32_t hmfai; > + uint8_t _reserved7[134 - 128]; /* 128-133 */ > + uint8_t byte_134[1]; > struct CPUEntry entries[0]; > } QEMU_PACKED ReadInfo; The size must be a multiple of 16. Can you add a reserved field to fill up until 144?
diff --git a/hw/s390x/sclp.c b/hw/s390x/sclp.c index 4510a80..183c627 100644 --- a/hw/s390x/sclp.c +++ b/hw/s390x/sclp.c @@ -73,6 +73,8 @@ static void read_SCP_info(SCLPDevice *sclp, SCCB *sccb) read_info->conf_char); s390_get_feat_block(S390_FEAT_TYPE_SCLP_CONF_CHAR_EXT, read_info->conf_char_ext); + /* Read Info byte 134 */ + s390_get_feat_block(S390_FEAT_TYPE_SCLP_BYTE_134, read_info->byte_134); read_info->facilities = cpu_to_be64(SCLP_HAS_CPU_INFO | SCLP_HAS_IOA_RECONFIG); diff --git a/include/hw/s390x/sclp.h b/include/hw/s390x/sclp.h index f9db243..eb12ba2 100644 --- a/include/hw/s390x/sclp.h +++ b/include/hw/s390x/sclp.h @@ -133,6 +133,8 @@ typedef struct ReadInfo { uint16_t highest_cpu; uint8_t _reserved5[124 - 122]; /* 122-123 */ uint32_t hmfai; + uint8_t _reserved7[134 - 128]; /* 128-133 */ + uint8_t byte_134[1]; struct CPUEntry entries[0]; } QEMU_PACKED ReadInfo; diff --git a/target/s390x/cpu.h b/target/s390x/cpu.h index 8c2320e..594b4a4 100644 --- a/target/s390x/cpu.h +++ b/target/s390x/cpu.h @@ -52,7 +52,7 @@ #define MMU_USER_IDX 0 -#define S390_MAX_CPUS 248 +#define S390_MAX_CPUS 247 typedef struct PSW { uint64_t mask; diff --git a/target/s390x/cpu_features.c b/target/s390x/cpu_features.c index 60cfeba..d05afa5 100644 --- a/target/s390x/cpu_features.c +++ b/target/s390x/cpu_features.c @@ -121,6 +121,9 @@ static const S390FeatDef s390_features[] = { FEAT_INIT("pfmfi", S390_FEAT_TYPE_SCLP_CONF_CHAR_EXT, 9, "SIE: PFMF interpretation facility"), FEAT_INIT("ibs", S390_FEAT_TYPE_SCLP_CONF_CHAR_EXT, 10, "SIE: Interlock-and-broadcast-suppression facility"), + /* SCLP SCCB Byte 134 */ + FEAT_INIT("diag318", S390_FEAT_TYPE_SCLP_BYTE_134, 0, "SIE: Diagnose 318"), + FEAT_INIT("sief2", S390_FEAT_TYPE_SCLP_CPU, 4, "SIE: interception format 2 (Virtual SIE)"), FEAT_INIT("skey", S390_FEAT_TYPE_SCLP_CPU, 5, "SIE: Storage-key facility"), FEAT_INIT("gpereh", S390_FEAT_TYPE_SCLP_CPU, 10, "SIE: Guest-PER enhancement facility"), diff --git a/target/s390x/cpu_features.h b/target/s390x/cpu_features.h index effe790..e7248df 100644 --- a/target/s390x/cpu_features.h +++ b/target/s390x/cpu_features.h @@ -23,6 +23,7 @@ typedef enum { S390_FEAT_TYPE_STFL, S390_FEAT_TYPE_SCLP_CONF_CHAR, S390_FEAT_TYPE_SCLP_CONF_CHAR_EXT, + S390_FEAT_TYPE_SCLP_BYTE_134, S390_FEAT_TYPE_SCLP_CPU, S390_FEAT_TYPE_MISC, S390_FEAT_TYPE_PLO, diff --git a/target/s390x/cpu_features_def.h b/target/s390x/cpu_features_def.h index 5fc7e7b..d99da1d 100644 --- a/target/s390x/cpu_features_def.h +++ b/target/s390x/cpu_features_def.h @@ -109,6 +109,9 @@ typedef enum { S390_FEAT_SIE_PFMFI, S390_FEAT_SIE_IBS, + /* Read Info Byte 134 */ + S390_FEAT_DIAG318, + /* Sclp Cpu */ S390_FEAT_SIE_F2, S390_FEAT_SIE_SKEY, diff --git a/target/s390x/gen-features.c b/target/s390x/gen-features.c index 70015ea..a3d1457 100644 --- a/target/s390x/gen-features.c +++ b/target/s390x/gen-features.c @@ -450,6 +450,7 @@ static uint16_t full_GEN12_GA1[] = { S390_FEAT_AP_QUERY_CONFIG_INFO, S390_FEAT_AP_FACILITIES_TEST, S390_FEAT_AP, + S390_FEAT_DIAG318, }; static uint16_t full_GEN12_GA2[] = { diff --git a/target/s390x/kvm.c b/target/s390x/kvm.c index 2ebf26a..3915e36 100644 --- a/target/s390x/kvm.c +++ b/target/s390x/kvm.c @@ -2142,6 +2142,7 @@ static int kvm_to_feat[][2] = { { KVM_S390_VM_CPU_FEAT_PFMFI, S390_FEAT_SIE_PFMFI}, { KVM_S390_VM_CPU_FEAT_SIGPIF, S390_FEAT_SIE_SIGPIF}, { KVM_S390_VM_CPU_FEAT_KSS, S390_FEAT_SIE_KSS}, + { KVM_S390_VM_CPU_FEAT_DIAG318, S390_FEAT_DIAG318}, }; static int query_cpu_feat(S390FeatBitmap features)
Diagnose 318 is a new z14.2 CPU feature. Since we are able to emulate it entirely via KVM, we can add guest support for earlier models. A new CPU feature for diagnose 318 (shortened to diag318) will be made available to guests starting with the zEC12-full CPU model. The z14.2 adds a new read SCP info byte (let's call it byte 134) to detect the availability of diag318. Because of this, we have room for one less VCPU and thus limit the max VPUs supported in a configuration to 247 (down from 248). Signed-off-by: Collin Walling <walling@linux.ibm.com>. --- hw/s390x/sclp.c | 2 ++ include/hw/s390x/sclp.h | 2 ++ target/s390x/cpu.h | 2 +- target/s390x/cpu_features.c | 3 +++ target/s390x/cpu_features.h | 1 + target/s390x/cpu_features_def.h | 3 +++ target/s390x/gen-features.c | 1 + target/s390x/kvm.c | 1 + 8 files changed, 14 insertions(+), 1 deletion(-)