Since Linux commit 00f918f61c56 ("RISC-V: KVM: Skeletal in-kernel AIA
irqchip support") checking KVM_CAP_IRQCHIP returns non-zero when the
RISC-V platform has AIA. The cap indicates KVM supports at least one
of the following ioctls:

  KVM_CREATE_IRQCHIP
  KVM_IRQ_LINE
  KVM_GET_IRQCHIP
  KVM_SET_IRQCHIP
  KVM_GET_LAPIC
  KVM_SET_LAPIC

but the cap doesn't imply that KVM must support any of those ioctls
in particular. However, QEMU was assuming the KVM_CREATE_IRQCHIP
ioctl was supported. Stop making that assumption by introducing a
KVM parameter that each architecture which supports KVM_CREATE_IRQCHIP
sets. Adding parameters isn't awesome, but given how the
KVM_CAP_IRQCHIP isn't very helpful on its own, we don't have a lot of
options.

Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
---

While this fixes booting guests on riscv KVM with AIA it's unlikely
to get merged before the QEMU support for KVM AIA[1] lands, which
would also fix the issue. I think this patch is still worth considering
though since QEMU's assumption is wrong.

[1] https://lore.kernel.org/all/20230714084429.22349-1-yongxuan.wang@sifive.com/

v2:
  - Move the s390x code to an s390x file. [Thomas]
  - Drop the KVM_CAP_IRQCHIP check from the top of kvm_irqchip_create(),
    as it's no longer necessary.

 accel/kvm/kvm-all.c    | 16 ++++------------
 include/sysemu/kvm.h   |  1 +
 target/arm/kvm.c       |  3 +++
 target/i386/kvm/kvm.c  |  2 ++
 target/s390x/kvm/kvm.c | 11 +++++++++++
 5 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index 373d876c0580..cddcb6eca641 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -86,6 +86,7 @@ struct KVMParkedVcpu {
 };
 
 KVMState *kvm_state;
+bool kvm_has_create_irqchip;
 bool kvm_kernel_irqchip;
 bool kvm_split_irqchip;
 bool kvm_async_interrupts_allowed;
@@ -2358,17 +2359,6 @@ static void kvm_irqchip_create(KVMState *s)
     int ret;
 
     assert(s->kernel_irqchip_split != ON_OFF_AUTO_AUTO);
-    if (kvm_check_extension(s, KVM_CAP_IRQCHIP)) {
-        ;
-    } else if (kvm_check_extension(s, KVM_CAP_S390_IRQCHIP)) {
-        ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
-        if (ret < 0) {
-            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
-            exit(1);
-        }
-    } else {
-        return;
-    }
 
     /* First probe and see if there's a arch-specific hook to create the
      * in-kernel irqchip for us */
@@ -2377,8 +2367,10 @@ static void kvm_irqchip_create(KVMState *s)
         if (s->kernel_irqchip_split == ON_OFF_AUTO_ON) {
             error_report("Split IRQ chip mode not supported.");
             exit(1);
-        } else {
+        } else if (kvm_has_create_irqchip) {
             ret = kvm_vm_ioctl(s, KVM_CREATE_IRQCHIP);
+        } else {
+            return;
         }
     }
     if (ret < 0) {
diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h
index 115f0cca79d1..84b1bb3dc91e 100644
--- a/include/sysemu/kvm.h
+++ b/include/sysemu/kvm.h
@@ -32,6 +32,7 @@
 #ifdef CONFIG_KVM_IS_POSSIBLE
 
 extern bool kvm_allowed;
+extern bool kvm_has_create_irqchip;
 extern bool kvm_kernel_irqchip;
 extern bool kvm_split_irqchip;
 extern bool kvm_async_interrupts_allowed;
diff --git a/target/arm/kvm.c b/target/arm/kvm.c
index b4c7654f4980..2fa87b495d68 100644
--- a/target/arm/kvm.c
+++ b/target/arm/kvm.c
@@ -250,6 +250,9 @@ int kvm_arm_get_max_vm_ipa_size(MachineState *ms, bool *fixed_ipa)
 int kvm_arch_init(MachineState *ms, KVMState *s)
 {
     int ret = 0;
+
+    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
+
     /* For ARM interrupt delivery is always asynchronous,
      * whether we are using an in-kernel VGIC or not.
      */
diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
index ebfaf3d24c79..6363e67f092d 100644
--- a/target/i386/kvm/kvm.c
+++ b/target/i386/kvm/kvm.c
@@ -2771,6 +2771,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
         }
     }
 
+    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
+
     return 0;
 }
 
diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
index a9e5880349d9..bcc735227f7d 100644
--- a/target/s390x/kvm/kvm.c
+++ b/target/s390x/kvm/kvm.c
@@ -391,6 +391,17 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
     }
 
     kvm_set_max_memslot_size(KVM_SLOT_MAX_BYTES);
+
+    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_S390_IRQCHIP);
+    if (kvm_has_create_irqchip) {
+        int ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
+
+        if (ret < 0) {
+            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
+            exit(1);
+        }
+    }
+
     return 0;
 }
 
-- 
2.41.0

Hi Paolo,

Is this good for 8.1?

Thanks,
drew


On Tue, Jul 25, 2023 at 02:26:02PM +0200, Andrew Jones wrote:
> Since Linux commit 00f918f61c56 ("RISC-V: KVM: Skeletal in-kernel AIA
> irqchip support") checking KVM_CAP_IRQCHIP returns non-zero when the
> RISC-V platform has AIA. The cap indicates KVM supports at least one
> of the following ioctls:
> 
>   KVM_CREATE_IRQCHIP
>   KVM_IRQ_LINE
>   KVM_GET_IRQCHIP
>   KVM_SET_IRQCHIP
>   KVM_GET_LAPIC
>   KVM_SET_LAPIC
> 
> but the cap doesn't imply that KVM must support any of those ioctls
> in particular. However, QEMU was assuming the KVM_CREATE_IRQCHIP
> ioctl was supported. Stop making that assumption by introducing a
> KVM parameter that each architecture which supports KVM_CREATE_IRQCHIP
> sets. Adding parameters isn't awesome, but given how the
> KVM_CAP_IRQCHIP isn't very helpful on its own, we don't have a lot of
> options.
> 
> Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
> ---
> 
> While this fixes booting guests on riscv KVM with AIA it's unlikely
> to get merged before the QEMU support for KVM AIA[1] lands, which
> would also fix the issue. I think this patch is still worth considering
> though since QEMU's assumption is wrong.
> 
> [1] https://lore.kernel.org/all/20230714084429.22349-1-yongxuan.wang@sifive.com/
> 
> v2:
>   - Move the s390x code to an s390x file. [Thomas]
>   - Drop the KVM_CAP_IRQCHIP check from the top of kvm_irqchip_create(),
>     as it's no longer necessary.
> 
>  accel/kvm/kvm-all.c    | 16 ++++------------
>  include/sysemu/kvm.h   |  1 +
>  target/arm/kvm.c       |  3 +++
>  target/i386/kvm/kvm.c  |  2 ++
>  target/s390x/kvm/kvm.c | 11 +++++++++++
>  5 files changed, 21 insertions(+), 12 deletions(-)
> 
> diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
> index 373d876c0580..cddcb6eca641 100644
> --- a/accel/kvm/kvm-all.c
> +++ b/accel/kvm/kvm-all.c
> @@ -86,6 +86,7 @@ struct KVMParkedVcpu {
>  };
>  
>  KVMState *kvm_state;
> +bool kvm_has_create_irqchip;
>  bool kvm_kernel_irqchip;
>  bool kvm_split_irqchip;
>  bool kvm_async_interrupts_allowed;
> @@ -2358,17 +2359,6 @@ static void kvm_irqchip_create(KVMState *s)
>      int ret;
>  
>      assert(s->kernel_irqchip_split != ON_OFF_AUTO_AUTO);
> -    if (kvm_check_extension(s, KVM_CAP_IRQCHIP)) {
> -        ;
> -    } else if (kvm_check_extension(s, KVM_CAP_S390_IRQCHIP)) {
> -        ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
> -        if (ret < 0) {
> -            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
> -            exit(1);
> -        }
> -    } else {
> -        return;
> -    }
>  
>      /* First probe and see if there's a arch-specific hook to create the
>       * in-kernel irqchip for us */
> @@ -2377,8 +2367,10 @@ static void kvm_irqchip_create(KVMState *s)
>          if (s->kernel_irqchip_split == ON_OFF_AUTO_ON) {
>              error_report("Split IRQ chip mode not supported.");
>              exit(1);
> -        } else {
> +        } else if (kvm_has_create_irqchip) {
>              ret = kvm_vm_ioctl(s, KVM_CREATE_IRQCHIP);
> +        } else {
> +            return;
>          }
>      }
>      if (ret < 0) {
> diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h
> index 115f0cca79d1..84b1bb3dc91e 100644
> --- a/include/sysemu/kvm.h
> +++ b/include/sysemu/kvm.h
> @@ -32,6 +32,7 @@
>  #ifdef CONFIG_KVM_IS_POSSIBLE
>  
>  extern bool kvm_allowed;
> +extern bool kvm_has_create_irqchip;
>  extern bool kvm_kernel_irqchip;
>  extern bool kvm_split_irqchip;
>  extern bool kvm_async_interrupts_allowed;
> diff --git a/target/arm/kvm.c b/target/arm/kvm.c
> index b4c7654f4980..2fa87b495d68 100644
> --- a/target/arm/kvm.c
> +++ b/target/arm/kvm.c
> @@ -250,6 +250,9 @@ int kvm_arm_get_max_vm_ipa_size(MachineState *ms, bool *fixed_ipa)
>  int kvm_arch_init(MachineState *ms, KVMState *s)
>  {
>      int ret = 0;
> +
> +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
> +
>      /* For ARM interrupt delivery is always asynchronous,
>       * whether we are using an in-kernel VGIC or not.
>       */
> diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
> index ebfaf3d24c79..6363e67f092d 100644
> --- a/target/i386/kvm/kvm.c
> +++ b/target/i386/kvm/kvm.c
> @@ -2771,6 +2771,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
>          }
>      }
>  
> +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
> +
>      return 0;
>  }
>  
> diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
> index a9e5880349d9..bcc735227f7d 100644
> --- a/target/s390x/kvm/kvm.c
> +++ b/target/s390x/kvm/kvm.c
> @@ -391,6 +391,17 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
>      }
>  
>      kvm_set_max_memslot_size(KVM_SLOT_MAX_BYTES);
> +
> +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_S390_IRQCHIP);
> +    if (kvm_has_create_irqchip) {
> +        int ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
> +
> +        if (ret < 0) {
> +            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
> +            exit(1);
> +        }
> +    }
> +
>      return 0;
>  }
>  
> -- 
> 2.41.0
>

On Thu, 10 Aug 2023 at 12:29, Andrew Jones <ajones@ventanamicro.com> wrote:
>
>
> Hi Paolo,
>
> Is this good for 8.1?

Is it fixing a regression since 8.0 ?

thanks
-- PMM

>
> Thanks,
> drew
>
>
> On Tue, Jul 25, 2023 at 02:26:02PM +0200, Andrew Jones wrote:
> > Since Linux commit 00f918f61c56 ("RISC-V: KVM: Skeletal in-kernel AIA
> > irqchip support") checking KVM_CAP_IRQCHIP returns non-zero when the
> > RISC-V platform has AIA. The cap indicates KVM supports at least one
> > of the following ioctls:
> >
> >   KVM_CREATE_IRQCHIP
> >   KVM_IRQ_LINE
> >   KVM_GET_IRQCHIP
> >   KVM_SET_IRQCHIP
> >   KVM_GET_LAPIC
> >   KVM_SET_LAPIC
> >
> > but the cap doesn't imply that KVM must support any of those ioctls
> > in particular. However, QEMU was assuming the KVM_CREATE_IRQCHIP
> > ioctl was supported. Stop making that assumption by introducing a
> > KVM parameter that each architecture which supports KVM_CREATE_IRQCHIP
> > sets. Adding parameters isn't awesome, but given how the
> > KVM_CAP_IRQCHIP isn't very helpful on its own, we don't have a lot of
> > options.
> >
> > Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
> > ---
> >
> > While this fixes booting guests on riscv KVM with AIA it's unlikely
> > to get merged before the QEMU support for KVM AIA[1] lands, which
> > would also fix the issue. I think this patch is still worth considering
> > though since QEMU's assumption is wrong.
> >
> > [1] https://lore.kernel.org/all/20230714084429.22349-1-yongxuan.wang@sifive.com/
> >
> > v2:
> >   - Move the s390x code to an s390x file. [Thomas]
> >   - Drop the KVM_CAP_IRQCHIP check from the top of kvm_irqchip_create(),
> >     as it's no longer necessary.
> >
> >  accel/kvm/kvm-all.c    | 16 ++++------------
> >  include/sysemu/kvm.h   |  1 +
> >  target/arm/kvm.c       |  3 +++
> >  target/i386/kvm/kvm.c  |  2 ++
> >  target/s390x/kvm/kvm.c | 11 +++++++++++
> >  5 files changed, 21 insertions(+), 12 deletions(-)
> >
> > diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
> > index 373d876c0580..cddcb6eca641 100644
> > --- a/accel/kvm/kvm-all.c
> > +++ b/accel/kvm/kvm-all.c
> > @@ -86,6 +86,7 @@ struct KVMParkedVcpu {
> >  };
> >
> >  KVMState *kvm_state;
> > +bool kvm_has_create_irqchip;
> >  bool kvm_kernel_irqchip;
> >  bool kvm_split_irqchip;
> >  bool kvm_async_interrupts_allowed;
> > @@ -2358,17 +2359,6 @@ static void kvm_irqchip_create(KVMState *s)
> >      int ret;
> >
> >      assert(s->kernel_irqchip_split != ON_OFF_AUTO_AUTO);
> > -    if (kvm_check_extension(s, KVM_CAP_IRQCHIP)) {
> > -        ;
> > -    } else if (kvm_check_extension(s, KVM_CAP_S390_IRQCHIP)) {
> > -        ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
> > -        if (ret < 0) {
> > -            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
> > -            exit(1);
> > -        }
> > -    } else {
> > -        return;
> > -    }
> >
> >      /* First probe and see if there's a arch-specific hook to create the
> >       * in-kernel irqchip for us */
> > @@ -2377,8 +2367,10 @@ static void kvm_irqchip_create(KVMState *s)
> >          if (s->kernel_irqchip_split == ON_OFF_AUTO_ON) {
> >              error_report("Split IRQ chip mode not supported.");
> >              exit(1);
> > -        } else {
> > +        } else if (kvm_has_create_irqchip) {
> >              ret = kvm_vm_ioctl(s, KVM_CREATE_IRQCHIP);
> > +        } else {
> > +            return;
> >          }
> >      }
> >      if (ret < 0) {
> > diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h
> > index 115f0cca79d1..84b1bb3dc91e 100644
> > --- a/include/sysemu/kvm.h
> > +++ b/include/sysemu/kvm.h
> > @@ -32,6 +32,7 @@
> >  #ifdef CONFIG_KVM_IS_POSSIBLE
> >
> >  extern bool kvm_allowed;
> > +extern bool kvm_has_create_irqchip;
> >  extern bool kvm_kernel_irqchip;
> >  extern bool kvm_split_irqchip;
> >  extern bool kvm_async_interrupts_allowed;
> > diff --git a/target/arm/kvm.c b/target/arm/kvm.c
> > index b4c7654f4980..2fa87b495d68 100644
> > --- a/target/arm/kvm.c
> > +++ b/target/arm/kvm.c
> > @@ -250,6 +250,9 @@ int kvm_arm_get_max_vm_ipa_size(MachineState *ms, bool *fixed_ipa)
> >  int kvm_arch_init(MachineState *ms, KVMState *s)
> >  {
> >      int ret = 0;
> > +
> > +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
> > +
> >      /* For ARM interrupt delivery is always asynchronous,
> >       * whether we are using an in-kernel VGIC or not.
> >       */
> > diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
> > index ebfaf3d24c79..6363e67f092d 100644
> > --- a/target/i386/kvm/kvm.c
> > +++ b/target/i386/kvm/kvm.c
> > @@ -2771,6 +2771,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
> >          }
> >      }
> >
> > +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
> > +
> >      return 0;
> >  }
> >
> > diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
> > index a9e5880349d9..bcc735227f7d 100644
> > --- a/target/s390x/kvm/kvm.c
> > +++ b/target/s390x/kvm/kvm.c
> > @@ -391,6 +391,17 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
> >      }
> >
> >      kvm_set_max_memslot_size(KVM_SLOT_MAX_BYTES);
> > +
> > +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_S390_IRQCHIP);
> > +    if (kvm_has_create_irqchip) {
> > +        int ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
> > +
> > +        if (ret < 0) {
> > +            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
> > +            exit(1);
> > +        }
> > +    }
> > +
> >      return 0;
> >  }
> >
> > --
> > 2.41.0
> >

On Thu, Aug 10, 2023 at 12:38:28PM +0100, Peter Maydell wrote:
> On Thu, 10 Aug 2023 at 12:29, Andrew Jones <ajones@ventanamicro.com> wrote:
> >
> >
> > Hi Paolo,
> >
> > Is this good for 8.1?
> 
> Is it fixing a regression since 8.0 ?

The fix is for running with KVM which includes 00f918f61c56, which is
v6.5-rc1 and later. All QEMU versions will fail to start, so this fix
isn't fixing a QEMU regression, but until this fix, or the KVM AIA
support series, is merged, then we won't have any QEMU version we can
use with later KVM.

Thanks,
drew

> 
> thanks
> -- PMM
> 
> >
> > Thanks,
> > drew
> >
> >
> > On Tue, Jul 25, 2023 at 02:26:02PM +0200, Andrew Jones wrote:
> > > Since Linux commit 00f918f61c56 ("RISC-V: KVM: Skeletal in-kernel AIA
> > > irqchip support") checking KVM_CAP_IRQCHIP returns non-zero when the
> > > RISC-V platform has AIA. The cap indicates KVM supports at least one
> > > of the following ioctls:
> > >
> > >   KVM_CREATE_IRQCHIP
> > >   KVM_IRQ_LINE
> > >   KVM_GET_IRQCHIP
> > >   KVM_SET_IRQCHIP
> > >   KVM_GET_LAPIC
> > >   KVM_SET_LAPIC
> > >
> > > but the cap doesn't imply that KVM must support any of those ioctls
> > > in particular. However, QEMU was assuming the KVM_CREATE_IRQCHIP
> > > ioctl was supported. Stop making that assumption by introducing a
> > > KVM parameter that each architecture which supports KVM_CREATE_IRQCHIP
> > > sets. Adding parameters isn't awesome, but given how the
> > > KVM_CAP_IRQCHIP isn't very helpful on its own, we don't have a lot of
> > > options.
> > >
> > > Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
> > > ---
> > >
> > > While this fixes booting guests on riscv KVM with AIA it's unlikely
> > > to get merged before the QEMU support for KVM AIA[1] lands, which
> > > would also fix the issue. I think this patch is still worth considering
> > > though since QEMU's assumption is wrong.
> > >
> > > [1] https://lore.kernel.org/all/20230714084429.22349-1-yongxuan.wang@sifive.com/
> > >
> > > v2:
> > >   - Move the s390x code to an s390x file. [Thomas]
> > >   - Drop the KVM_CAP_IRQCHIP check from the top of kvm_irqchip_create(),
> > >     as it's no longer necessary.
> > >
> > >  accel/kvm/kvm-all.c    | 16 ++++------------
> > >  include/sysemu/kvm.h   |  1 +
> > >  target/arm/kvm.c       |  3 +++
> > >  target/i386/kvm/kvm.c  |  2 ++
> > >  target/s390x/kvm/kvm.c | 11 +++++++++++
> > >  5 files changed, 21 insertions(+), 12 deletions(-)
> > >
> > > diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
> > > index 373d876c0580..cddcb6eca641 100644
> > > --- a/accel/kvm/kvm-all.c
> > > +++ b/accel/kvm/kvm-all.c
> > > @@ -86,6 +86,7 @@ struct KVMParkedVcpu {
> > >  };
> > >
> > >  KVMState *kvm_state;
> > > +bool kvm_has_create_irqchip;
> > >  bool kvm_kernel_irqchip;
> > >  bool kvm_split_irqchip;
> > >  bool kvm_async_interrupts_allowed;
> > > @@ -2358,17 +2359,6 @@ static void kvm_irqchip_create(KVMState *s)
> > >      int ret;
> > >
> > >      assert(s->kernel_irqchip_split != ON_OFF_AUTO_AUTO);
> > > -    if (kvm_check_extension(s, KVM_CAP_IRQCHIP)) {
> > > -        ;
> > > -    } else if (kvm_check_extension(s, KVM_CAP_S390_IRQCHIP)) {
> > > -        ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
> > > -        if (ret < 0) {
> > > -            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
> > > -            exit(1);
> > > -        }
> > > -    } else {
> > > -        return;
> > > -    }
> > >
> > >      /* First probe and see if there's a arch-specific hook to create the
> > >       * in-kernel irqchip for us */
> > > @@ -2377,8 +2367,10 @@ static void kvm_irqchip_create(KVMState *s)
> > >          if (s->kernel_irqchip_split == ON_OFF_AUTO_ON) {
> > >              error_report("Split IRQ chip mode not supported.");
> > >              exit(1);
> > > -        } else {
> > > +        } else if (kvm_has_create_irqchip) {
> > >              ret = kvm_vm_ioctl(s, KVM_CREATE_IRQCHIP);
> > > +        } else {
> > > +            return;
> > >          }
> > >      }
> > >      if (ret < 0) {
> > > diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h
> > > index 115f0cca79d1..84b1bb3dc91e 100644
> > > --- a/include/sysemu/kvm.h
> > > +++ b/include/sysemu/kvm.h
> > > @@ -32,6 +32,7 @@
> > >  #ifdef CONFIG_KVM_IS_POSSIBLE
> > >
> > >  extern bool kvm_allowed;
> > > +extern bool kvm_has_create_irqchip;
> > >  extern bool kvm_kernel_irqchip;
> > >  extern bool kvm_split_irqchip;
> > >  extern bool kvm_async_interrupts_allowed;
> > > diff --git a/target/arm/kvm.c b/target/arm/kvm.c
> > > index b4c7654f4980..2fa87b495d68 100644
> > > --- a/target/arm/kvm.c
> > > +++ b/target/arm/kvm.c
> > > @@ -250,6 +250,9 @@ int kvm_arm_get_max_vm_ipa_size(MachineState *ms, bool *fixed_ipa)
> > >  int kvm_arch_init(MachineState *ms, KVMState *s)
> > >  {
> > >      int ret = 0;
> > > +
> > > +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
> > > +
> > >      /* For ARM interrupt delivery is always asynchronous,
> > >       * whether we are using an in-kernel VGIC or not.
> > >       */
> > > diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
> > > index ebfaf3d24c79..6363e67f092d 100644
> > > --- a/target/i386/kvm/kvm.c
> > > +++ b/target/i386/kvm/kvm.c
> > > @@ -2771,6 +2771,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
> > >          }
> > >      }
> > >
> > > +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
> > > +
> > >      return 0;
> > >  }
> > >
> > > diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
> > > index a9e5880349d9..bcc735227f7d 100644
> > > --- a/target/s390x/kvm/kvm.c
> > > +++ b/target/s390x/kvm/kvm.c
> > > @@ -391,6 +391,17 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
> > >      }
> > >
> > >      kvm_set_max_memslot_size(KVM_SLOT_MAX_BYTES);
> > > +
> > > +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_S390_IRQCHIP);
> > > +    if (kvm_has_create_irqchip) {
> > > +        int ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
> > > +
> > > +        if (ret < 0) {
> > > +            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
> > > +            exit(1);
> > > +        }
> > > +    }
> > > +
> > >      return 0;
> > >  }
> > >
> > > --
> > > 2.41.0
> > >

On 7/25/23 09:26, Andrew Jones wrote:
> Since Linux commit 00f918f61c56 ("RISC-V: KVM: Skeletal in-kernel AIA
> irqchip support") checking KVM_CAP_IRQCHIP returns non-zero when the
> RISC-V platform has AIA. The cap indicates KVM supports at least one
> of the following ioctls:
> 
>    KVM_CREATE_IRQCHIP
>    KVM_IRQ_LINE
>    KVM_GET_IRQCHIP
>    KVM_SET_IRQCHIP
>    KVM_GET_LAPIC
>    KVM_SET_LAPIC
> 
> but the cap doesn't imply that KVM must support any of those ioctls
> in particular. However, QEMU was assuming the KVM_CREATE_IRQCHIP
> ioctl was supported. Stop making that assumption by introducing a
> KVM parameter that each architecture which supports KVM_CREATE_IRQCHIP
> sets. Adding parameters isn't awesome, but given how the
> KVM_CAP_IRQCHIP isn't very helpful on its own, we don't have a lot of
> options.
> 
> Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
> ---

Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>

> 
> While this fixes booting guests on riscv KVM with AIA it's unlikely
> to get merged before the QEMU support for KVM AIA[1] lands, which
> would also fix the issue. I think this patch is still worth considering
> though since QEMU's assumption is wrong.
> 
> [1] https://lore.kernel.org/all/20230714084429.22349-1-yongxuan.wang@sifive.com/
> 
> v2:
>    - Move the s390x code to an s390x file. [Thomas]
>    - Drop the KVM_CAP_IRQCHIP check from the top of kvm_irqchip_create(),
>      as it's no longer necessary.
> 
>   accel/kvm/kvm-all.c    | 16 ++++------------
>   include/sysemu/kvm.h   |  1 +
>   target/arm/kvm.c       |  3 +++
>   target/i386/kvm/kvm.c  |  2 ++
>   target/s390x/kvm/kvm.c | 11 +++++++++++
>   5 files changed, 21 insertions(+), 12 deletions(-)
> 
> diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
> index 373d876c0580..cddcb6eca641 100644
> --- a/accel/kvm/kvm-all.c
> +++ b/accel/kvm/kvm-all.c
> @@ -86,6 +86,7 @@ struct KVMParkedVcpu {
>   };
>   
>   KVMState *kvm_state;
> +bool kvm_has_create_irqchip;
>   bool kvm_kernel_irqchip;
>   bool kvm_split_irqchip;
>   bool kvm_async_interrupts_allowed;
> @@ -2358,17 +2359,6 @@ static void kvm_irqchip_create(KVMState *s)
>       int ret;
>   
>       assert(s->kernel_irqchip_split != ON_OFF_AUTO_AUTO);
> -    if (kvm_check_extension(s, KVM_CAP_IRQCHIP)) {
> -        ;
> -    } else if (kvm_check_extension(s, KVM_CAP_S390_IRQCHIP)) {
> -        ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
> -        if (ret < 0) {
> -            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
> -            exit(1);
> -        }
> -    } else {
> -        return;
> -    }
>   
>       /* First probe and see if there's a arch-specific hook to create the
>        * in-kernel irqchip for us */
> @@ -2377,8 +2367,10 @@ static void kvm_irqchip_create(KVMState *s)
>           if (s->kernel_irqchip_split == ON_OFF_AUTO_ON) {
>               error_report("Split IRQ chip mode not supported.");
>               exit(1);
> -        } else {
> +        } else if (kvm_has_create_irqchip) {
>               ret = kvm_vm_ioctl(s, KVM_CREATE_IRQCHIP);
> +        } else {
> +            return;
>           }
>       }
>       if (ret < 0) {
> diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h
> index 115f0cca79d1..84b1bb3dc91e 100644
> --- a/include/sysemu/kvm.h
> +++ b/include/sysemu/kvm.h
> @@ -32,6 +32,7 @@
>   #ifdef CONFIG_KVM_IS_POSSIBLE
>   
>   extern bool kvm_allowed;
> +extern bool kvm_has_create_irqchip;
>   extern bool kvm_kernel_irqchip;
>   extern bool kvm_split_irqchip;
>   extern bool kvm_async_interrupts_allowed;
> diff --git a/target/arm/kvm.c b/target/arm/kvm.c
> index b4c7654f4980..2fa87b495d68 100644
> --- a/target/arm/kvm.c
> +++ b/target/arm/kvm.c
> @@ -250,6 +250,9 @@ int kvm_arm_get_max_vm_ipa_size(MachineState *ms, bool *fixed_ipa)
>   int kvm_arch_init(MachineState *ms, KVMState *s)
>   {
>       int ret = 0;
> +
> +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
> +
>       /* For ARM interrupt delivery is always asynchronous,
>        * whether we are using an in-kernel VGIC or not.
>        */
> diff --git a/target/i386/kvm/kvm.c b/target/i386/kvm/kvm.c
> index ebfaf3d24c79..6363e67f092d 100644
> --- a/target/i386/kvm/kvm.c
> +++ b/target/i386/kvm/kvm.c
> @@ -2771,6 +2771,8 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
>           }
>       }
>   
> +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_IRQCHIP);
> +
>       return 0;
>   }
>   
> diff --git a/target/s390x/kvm/kvm.c b/target/s390x/kvm/kvm.c
> index a9e5880349d9..bcc735227f7d 100644
> --- a/target/s390x/kvm/kvm.c
> +++ b/target/s390x/kvm/kvm.c
> @@ -391,6 +391,17 @@ int kvm_arch_init(MachineState *ms, KVMState *s)
>       }
>   
>       kvm_set_max_memslot_size(KVM_SLOT_MAX_BYTES);
> +
> +    kvm_has_create_irqchip = kvm_check_extension(s, KVM_CAP_S390_IRQCHIP);
> +    if (kvm_has_create_irqchip) {
> +        int ret = kvm_vm_enable_cap(s, KVM_CAP_S390_IRQCHIP, 0);
> +
> +        if (ret < 0) {
> +            fprintf(stderr, "Enable kernel irqchip failed: %s\n", strerror(-ret));
> +            exit(1);
> +        }
> +    }
> +
>       return 0;
>   }
>

On 25/07/2023 14.26, Andrew Jones wrote:
> Since Linux commit 00f918f61c56 ("RISC-V: KVM: Skeletal in-kernel AIA
> irqchip support") checking KVM_CAP_IRQCHIP returns non-zero when the
> RISC-V platform has AIA. The cap indicates KVM supports at least one
> of the following ioctls:
> 
>    KVM_CREATE_IRQCHIP
>    KVM_IRQ_LINE
>    KVM_GET_IRQCHIP
>    KVM_SET_IRQCHIP
>    KVM_GET_LAPIC
>    KVM_SET_LAPIC
> 
> but the cap doesn't imply that KVM must support any of those ioctls
> in particular. However, QEMU was assuming the KVM_CREATE_IRQCHIP
> ioctl was supported. Stop making that assumption by introducing a
> KVM parameter that each architecture which supports KVM_CREATE_IRQCHIP
> sets. Adding parameters isn't awesome, but given how the
> KVM_CAP_IRQCHIP isn't very helpful on its own, we don't have a lot of
> options.
> 
> Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
> ---
> 
> While this fixes booting guests on riscv KVM with AIA it's unlikely
> to get merged before the QEMU support for KVM AIA[1] lands, which
> would also fix the issue. I think this patch is still worth considering
> though since QEMU's assumption is wrong.
> 
> [1] https://lore.kernel.org/all/20230714084429.22349-1-yongxuan.wang@sifive.com/
> 
> v2:
>    - Move the s390x code to an s390x file. [Thomas]
>    - Drop the KVM_CAP_IRQCHIP check from the top of kvm_irqchip_create(),
>      as it's no longer necessary.

Looks good now!

Reviewed-by: Thomas Huth <thuth@redhat.com>

On 25/7/23 14:26, Andrew Jones wrote:
> Since Linux commit 00f918f61c56 ("RISC-V: KVM: Skeletal in-kernel AIA
> irqchip support") checking KVM_CAP_IRQCHIP returns non-zero when the
> RISC-V platform has AIA. The cap indicates KVM supports at least one
> of the following ioctls:
> 
>    KVM_CREATE_IRQCHIP
>    KVM_IRQ_LINE
>    KVM_GET_IRQCHIP
>    KVM_SET_IRQCHIP
>    KVM_GET_LAPIC
>    KVM_SET_LAPIC
> 
> but the cap doesn't imply that KVM must support any of those ioctls
> in particular. However, QEMU was assuming the KVM_CREATE_IRQCHIP
> ioctl was supported. Stop making that assumption by introducing a
> KVM parameter that each architecture which supports KVM_CREATE_IRQCHIP
> sets. Adding parameters isn't awesome, but given how the
> KVM_CAP_IRQCHIP isn't very helpful on its own, we don't have a lot of
> options.
> 
> Signed-off-by: Andrew Jones <ajones@ventanamicro.com>
> ---
> 
> While this fixes booting guests on riscv KVM with AIA it's unlikely
> to get merged before the QEMU support for KVM AIA[1] lands, which
> would also fix the issue. I think this patch is still worth considering
> though since QEMU's assumption is wrong.
> 
> [1] https://lore.kernel.org/all/20230714084429.22349-1-yongxuan.wang@sifive.com/
> 
> v2:
>    - Move the s390x code to an s390x file. [Thomas]
>    - Drop the KVM_CAP_IRQCHIP check from the top of kvm_irqchip_create(),
>      as it's no longer necessary.
> 
>   accel/kvm/kvm-all.c    | 16 ++++------------
>   include/sysemu/kvm.h   |  1 +
>   target/arm/kvm.c       |  3 +++
>   target/i386/kvm/kvm.c  |  2 ++
>   target/s390x/kvm/kvm.c | 11 +++++++++++
>   5 files changed, 21 insertions(+), 12 deletions(-)

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>