[PATCH 0/2] kvm: fix KVM_MAX_VCPU_ID handling

[PATCH 0/2] kvm: fix KVM_MAX_VCPU_ID handling

[Juergen Gross]

Revert commit 76b4f357d0e7d8f6f00 which was based on wrong reasoning
and rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS in order to avoid the
same issue in future.

Juergen Gross (2):
  x86/kvm: revert commit 76b4f357d0e7d8f6f00
  kvm: rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS

 Documentation/virt/kvm/devices/xics.rst            | 2 +-
 Documentation/virt/kvm/devices/xive.rst            | 2 +-
 arch/mips/kvm/mips.c                               | 2 +-
 arch/powerpc/include/asm/kvm_book3s.h              | 2 +-
 arch/powerpc/include/asm/kvm_host.h                | 4 ++--
 arch/powerpc/kvm/book3s_xive.c                     | 2 +-
 arch/powerpc/kvm/powerpc.c                         | 2 +-
 arch/x86/include/asm/kvm_host.h                    | 2 +-
 arch/x86/kvm/ioapic.c                              | 2 +-
 arch/x86/kvm/ioapic.h                              | 4 ++--
 arch/x86/kvm/x86.c                                 | 2 +-
 include/linux/kvm_host.h                           | 4 ++--
 tools/testing/selftests/kvm/kvm_create_max_vcpus.c | 2 +-
 virt/kvm/kvm_main.c                                | 2 +-
 14 files changed, 17 insertions(+), 17 deletions(-)

-- 
2.26.2

[PATCH 2/2] kvm: rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS

[Juergen Gross]

KVM_MAX_VCPU_ID is not specifying the highest allowed vcpu-id, but the
number of allowed vcpu-ids. This has already led to confusion, so
rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS to make its semantics more
clear

Suggested-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
---
 Documentation/virt/kvm/devices/xics.rst            | 2 +-
 Documentation/virt/kvm/devices/xive.rst            | 2 +-
 arch/mips/kvm/mips.c                               | 2 +-
 arch/powerpc/include/asm/kvm_book3s.h              | 2 +-
 arch/powerpc/include/asm/kvm_host.h                | 4 ++--
 arch/powerpc/kvm/book3s_xive.c                     | 2 +-
 arch/powerpc/kvm/powerpc.c                         | 2 +-
 arch/x86/include/asm/kvm_host.h                    | 2 +-
 arch/x86/kvm/ioapic.c                              | 2 +-
 arch/x86/kvm/ioapic.h                              | 4 ++--
 arch/x86/kvm/x86.c                                 | 2 +-
 include/linux/kvm_host.h                           | 4 ++--
 tools/testing/selftests/kvm/kvm_create_max_vcpus.c | 2 +-
 virt/kvm/kvm_main.c                                | 2 +-
 14 files changed, 17 insertions(+), 17 deletions(-)

diff --git a/Documentation/virt/kvm/devices/xics.rst b/Documentation/virt/kvm/devices/xics.rst
index 2d6927e0b776..bf32c77174ab 100644
--- a/Documentation/virt/kvm/devices/xics.rst
+++ b/Documentation/virt/kvm/devices/xics.rst
@@ -22,7 +22,7 @@ Groups:
   Errors:
 
     =======  ==========================================
-    -EINVAL  Value greater than KVM_MAX_VCPU_ID.
+    -EINVAL  Value greater than KVM_MAX_VCPU_IDS.
     -EFAULT  Invalid user pointer for attr->addr.
     -EBUSY   A vcpu is already connected to the device.
     =======  ==========================================
diff --git a/Documentation/virt/kvm/devices/xive.rst b/Documentation/virt/kvm/devices/xive.rst
index 8bdf3dc38f01..8b5e7b40bdf8 100644
--- a/Documentation/virt/kvm/devices/xive.rst
+++ b/Documentation/virt/kvm/devices/xive.rst
@@ -91,7 +91,7 @@ the legacy interrupt mode, referred as XICS (POWER7/8).
     Errors:
 
       =======  ==========================================
-      -EINVAL  Value greater than KVM_MAX_VCPU_ID.
+      -EINVAL  Value greater than KVM_MAX_VCPU_IDS.
       -EFAULT  Invalid user pointer for attr->addr.
       -EBUSY   A vCPU is already connected to the device.
       =======  ==========================================
diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c
index 75c6f264c626..562aa878b266 100644
--- a/arch/mips/kvm/mips.c
+++ b/arch/mips/kvm/mips.c
@@ -1073,7 +1073,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 		r = KVM_MAX_VCPUS;
 		break;
 	case KVM_CAP_MAX_VCPU_ID:
-		r = KVM_MAX_VCPU_ID;
+		r = KVM_MAX_VCPU_IDS;
 		break;
 	case KVM_CAP_MIPS_FPU:
 		/* We don't handle systems with inconsistent cpu_has_fpu */
diff --git a/arch/powerpc/include/asm/kvm_book3s.h b/arch/powerpc/include/asm/kvm_book3s.h
index caaa0f592d8e..3d31f2c59e43 100644
--- a/arch/powerpc/include/asm/kvm_book3s.h
+++ b/arch/powerpc/include/asm/kvm_book3s.h
@@ -434,7 +434,7 @@ extern int kvmppc_h_logical_ci_store(struct kvm_vcpu *vcpu);
 #define SPLIT_HACK_OFFS			0xfb000000
 
 /*
- * This packs a VCPU ID from the [0..KVM_MAX_VCPU_ID) space down to the
+ * This packs a VCPU ID from the [0..KVM_MAX_VCPU_IDS) space down to the
  * [0..KVM_MAX_VCPUS) space, using knowledge of the guest's core stride
  * (but not its actual threading mode, which is not available) to avoid
  * collisions.
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index 080a7feb7731..59cb38b04ede 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -33,11 +33,11 @@
 
 #ifdef CONFIG_KVM_BOOK3S_HV_POSSIBLE
 #include <asm/kvm_book3s_asm.h>		/* for MAX_SMT_THREADS */
-#define KVM_MAX_VCPU_ID		(MAX_SMT_THREADS * KVM_MAX_VCORES)
+#define KVM_MAX_VCPU_IDS	(MAX_SMT_THREADS * KVM_MAX_VCORES)
 #define KVM_MAX_NESTED_GUESTS	KVMPPC_NR_LPIDS
 
 #else
-#define KVM_MAX_VCPU_ID		KVM_MAX_VCPUS
+#define KVM_MAX_VCPU_IDS	KVM_MAX_VCPUS
 #endif /* CONFIG_KVM_BOOK3S_HV_POSSIBLE */
 
 #define __KVM_HAVE_ARCH_INTC_INITIALIZED
diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c
index a18db9e16ea4..225008882958 100644
--- a/arch/powerpc/kvm/book3s_xive.c
+++ b/arch/powerpc/kvm/book3s_xive.c
@@ -1928,7 +1928,7 @@ int kvmppc_xive_set_nr_servers(struct kvmppc_xive *xive, u64 addr)
 
 	pr_devel("%s nr_servers=%u\n", __func__, nr_servers);
 
-	if (!nr_servers || nr_servers > KVM_MAX_VCPU_ID)
+	if (!nr_servers || nr_servers > KVM_MAX_VCPU_IDS)
 		return -EINVAL;
 
 	mutex_lock(&xive->lock);
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index b4e6f70b97b9..8ab90ce8738f 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -649,7 +649,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 		r = KVM_MAX_VCPUS;
 		break;
 	case KVM_CAP_MAX_VCPU_ID:
-		r = KVM_MAX_VCPU_ID;
+		r = KVM_MAX_VCPU_IDS;
 		break;
 #ifdef CONFIG_PPC_BOOK3S_64
 	case KVM_CAP_PPC_GET_SMMU_INFO:
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index f8f48a7ec577..261b7d857fc0 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -50,7 +50,7 @@
  * so ratio of 4 should be enough.
  */
 #define KVM_VCPU_ID_RATIO 4
-#define KVM_MAX_VCPU_ID (KVM_MAX_VCPUS * KVM_VCPU_ID_RATIO)
+#define KVM_MAX_VCPU_IDS (KVM_MAX_VCPUS * KVM_VCPU_ID_RATIO)
 
 /* memory slots that are not exposed to userspace */
 #define KVM_PRIVATE_MEM_SLOTS 3
diff --git a/arch/x86/kvm/ioapic.c b/arch/x86/kvm/ioapic.c
index 698969e18fe3..3358123be946 100644
--- a/arch/x86/kvm/ioapic.c
+++ b/arch/x86/kvm/ioapic.c
@@ -96,7 +96,7 @@ static unsigned long ioapic_read_indirect(struct kvm_ioapic *ioapic,
 static void rtc_irq_eoi_tracking_reset(struct kvm_ioapic *ioapic)
 {
 	ioapic->rtc_status.pending_eoi = 0;
-	bitmap_zero(ioapic->rtc_status.dest_map.map, KVM_MAX_VCPU_ID);
+	bitmap_zero(ioapic->rtc_status.dest_map.map, KVM_MAX_VCPU_IDS);
 }
 
 static void kvm_rtc_eoi_tracking_restore_all(struct kvm_ioapic *ioapic);
diff --git a/arch/x86/kvm/ioapic.h b/arch/x86/kvm/ioapic.h
index 27e61ff3ac3e..e66e620c3bed 100644
--- a/arch/x86/kvm/ioapic.h
+++ b/arch/x86/kvm/ioapic.h
@@ -39,13 +39,13 @@ struct kvm_vcpu;
 
 struct dest_map {
 	/* vcpu bitmap where IRQ has been sent */
-	DECLARE_BITMAP(map, KVM_MAX_VCPU_ID);
+	DECLARE_BITMAP(map, KVM_MAX_VCPU_IDS);
 
 	/*
 	 * Vector sent to a given vcpu, only valid when
 	 * the vcpu's bit in map is set
 	 */
-	u8 vectors[KVM_MAX_VCPU_ID];
+	u8 vectors[KVM_MAX_VCPU_IDS];
 };
 
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 28ef14155726..fec43997522b 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -4070,7 +4070,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 		r = KVM_MAX_VCPUS;
 		break;
 	case KVM_CAP_MAX_VCPU_ID:
-		r = KVM_MAX_VCPU_ID;
+		r = KVM_MAX_VCPU_IDS;
 		break;
 	case KVM_CAP_PV_MMU:	/* obsolete */
 		r = 0;
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 041ca7f15ea4..d2b405170f6e 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -39,8 +39,8 @@
 #include <asm/kvm_host.h>
 #include <linux/kvm_dirty_ring.h>
 
-#ifndef KVM_MAX_VCPU_ID
-#define KVM_MAX_VCPU_ID KVM_MAX_VCPUS
+#ifndef KVM_MAX_VCPU_IDS
+#define KVM_MAX_VCPU_IDS KVM_MAX_VCPUS
 #endif
 
 /*
diff --git a/tools/testing/selftests/kvm/kvm_create_max_vcpus.c b/tools/testing/selftests/kvm/kvm_create_max_vcpus.c
index 0299cd81b8ba..f968dfd4ee88 100644
--- a/tools/testing/selftests/kvm/kvm_create_max_vcpus.c
+++ b/tools/testing/selftests/kvm/kvm_create_max_vcpus.c
@@ -53,7 +53,7 @@ int main(int argc, char *argv[])
 		kvm_max_vcpu_id = kvm_max_vcpus;
 
 	TEST_ASSERT(kvm_max_vcpu_id >= kvm_max_vcpus,
-		    "KVM_MAX_VCPU_ID (%d) must be at least as large as KVM_MAX_VCPUS (%d).",
+		    "KVM_MAX_VCPU_IDS (%d) must be at least as large as KVM_MAX_VCPUS (%d).",
 		    kvm_max_vcpu_id, kvm_max_vcpus);
 
 	test_vcpu_creation(0, kvm_max_vcpus);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 439d3b4cd1a9..62fcbc897e1c 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -3557,7 +3557,7 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
 	struct kvm_vcpu *vcpu;
 	struct page *page;
 
-	if (id >= KVM_MAX_VCPU_ID)
+	if (id >= KVM_MAX_VCPU_IDS)
 		return -EINVAL;
 
 	mutex_lock(&kvm->lock);
-- 
2.26.2

Re: [PATCH 2/2] kvm: rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS

[Sean Christopherson]

On Mon, Sep 13, 2021, Juergen Gross wrote:
> KVM_MAX_VCPU_ID is not specifying the highest allowed vcpu-id, but the
> number of allowed vcpu-ids. This has already led to confusion, so
> rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS to make its semantics more
> clear

My hesitation with this rename is that the max _number_ of IDs is not the same
thing as the max allowed ID.  E.g. on x86, given a capability that enumerates the
max number of IDs, I would expect to be able to create vCPUs with arbitrary 32-bit
x2APIC IDs so long as the total number of IDs is below the max.

Re: [PATCH 2/2] kvm: rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS

[Eduardo Habkost]

On Mon, Sep 13, 2021 at 12:24 PM Sean Christopherson <seanjc@google.com> wrote:
>
> On Mon, Sep 13, 2021, Juergen Gross wrote:
> > KVM_MAX_VCPU_ID is not specifying the highest allowed vcpu-id, but the
> > number of allowed vcpu-ids. This has already led to confusion, so
> > rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS to make its semantics more
> > clear
>
> My hesitation with this rename is that the max _number_ of IDs is not the same
> thing as the max allowed ID.  E.g. on x86, given a capability that enumerates the
> max number of IDs, I would expect to be able to create vCPUs with arbitrary 32-bit
> x2APIC IDs so long as the total number of IDs is below the max.
>

What name would you suggest instead? KVM_VCPU_ID_LIMIT, maybe?

I'm assuming we are not going to redefine KVM_MAX_VCPU_ID to be an
inclusive limit.

--
Eduardo

Re: [PATCH 2/2] kvm: rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS

[Sean Christopherson]

On Mon, Sep 13, 2021, Eduardo Habkost wrote:
> On Mon, Sep 13, 2021 at 12:24 PM Sean Christopherson <seanjc@google.com> wrote:
> >
> > On Mon, Sep 13, 2021, Juergen Gross wrote:
> > > KVM_MAX_VCPU_ID is not specifying the highest allowed vcpu-id, but the
> > > number of allowed vcpu-ids. This has already led to confusion, so
> > > rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS to make its semantics more
> > > clear
> >
> > My hesitation with this rename is that the max _number_ of IDs is not the same
> > thing as the max allowed ID.  E.g. on x86, given a capability that enumerates the
> > max number of IDs, I would expect to be able to create vCPUs with arbitrary 32-bit
> > x2APIC IDs so long as the total number of IDs is below the max.
> >
> 
> What name would you suggest instead? KVM_VCPU_ID_LIMIT, maybe?
> 
> I'm assuming we are not going to redefine KVM_MAX_VCPU_ID to be an
> inclusive limit.

Heh, I haven't been able to come up with one, which is why I suggested the route
of making it an inclusive value internally within KVM.

Re: [PATCH 0/2] kvm: fix KVM_MAX_VCPU_ID handling

[Paolo Bonzini]

On 13/09/21 15:57, Juergen Gross wrote:
> Revert commit 76b4f357d0e7d8f6f00 which was based on wrong reasoning
> and rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS in order to avoid the
> same issue in future.
> 
> Juergen Gross (2):
>    x86/kvm: revert commit 76b4f357d0e7d8f6f00
>    kvm: rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS
> 
>   Documentation/virt/kvm/devices/xics.rst            | 2 +-
>   Documentation/virt/kvm/devices/xive.rst            | 2 +-
>   arch/mips/kvm/mips.c                               | 2 +-
>   arch/powerpc/include/asm/kvm_book3s.h              | 2 +-
>   arch/powerpc/include/asm/kvm_host.h                | 4 ++--
>   arch/powerpc/kvm/book3s_xive.c                     | 2 +-
>   arch/powerpc/kvm/powerpc.c                         | 2 +-
>   arch/x86/include/asm/kvm_host.h                    | 2 +-
>   arch/x86/kvm/ioapic.c                              | 2 +-
>   arch/x86/kvm/ioapic.h                              | 4 ++--
>   arch/x86/kvm/x86.c                                 | 2 +-
>   include/linux/kvm_host.h                           | 4 ++--
>   tools/testing/selftests/kvm/kvm_create_max_vcpus.c | 2 +-
>   virt/kvm/kvm_main.c                                | 2 +-
>   14 files changed, 17 insertions(+), 17 deletions(-)
> 

Queued, thanks.

Paolo

Re: [PATCH 2/2] kvm: rename KVM_MAX_VCPU_ID to, KVM_MAX_VCPU_IDS

[Christian Zigotzky]

Hello Juergen,
Hello All,

Since the RC1 of kernel 5.13, -smp 2 and -smp 4 don't work with a 
virtual e5500 QEMU KVM-HV machine anymore. [1]
I see in the serial console, that the uImage doesn't load. I use the 
following QEMU command for booting:

qemu-system-ppc64 -M ppce500 -cpu e5500 -enable-kvm -m 1024 -kernel 
uImage -drive format=raw,file=MintPPC32-X5000.img,index=0,if=virtio 
-netdev user,id=mynet0 -device virtio-net,netdev=mynet0 -append "rw 
root=/dev/vda" -device virtio-vga -device virtio-mouse-pci -device 
virtio-keyboard-pci -device pci-ohci,id=newusb -device 
usb-audio,bus=newusb.0 -smp 4

The kernels boot without KVM-HV.

Summary for KVM-HV:

-smp 1 -> works
-smp 2 -> doesn't work
-smp 3 -> works
-smp 4 -> doesn't work

I used -smp 4 before the RC1 of kernel 5.13 because my FSL P5040 BookE 
machine [2] has 4 cores.

Does this patch solve this issue? [3]

Thanks,
Christian

[1] https://lists.ozlabs.org/pipermail/linuxppc-dev/2021-May/229103.html
[2] http://wiki.amiga.org/index.php?title=X5000
[3] 
https://lists.ozlabs.org/pipermail/linuxppc-dev/2021-September/234152.html

Re: [PATCH 2/2] kvm: rename KVM_MAX_VCPU_ID to, KVM_MAX_VCPU_IDS

[Christian Zigotzky]

On 14 September 2021 at 05:59 pm, Christian Zigotzky wrote:
> Hello Juergen,
> Hello All,
>
> Since the RC1 of kernel 5.13, -smp 2 and -smp 4 don't work with a 
> virtual e5500 QEMU KVM-HV machine anymore. [1]
> I see in the serial console, that the uImage doesn't load. I use the 
> following QEMU command for booting:
>
> qemu-system-ppc64 -M ppce500 -cpu e5500 -enable-kvm -m 1024 -kernel 
> uImage -drive format=raw,file=MintPPC32-X5000.img,index=0,if=virtio 
> -netdev user,id=mynet0 -device virtio-net,netdev=mynet0 -append "rw 
> root=/dev/vda" -device virtio-vga -device virtio-mouse-pci -device 
> virtio-keyboard-pci -device pci-ohci,id=newusb -device 
> usb-audio,bus=newusb.0 -smp 4
>
> The kernels boot without KVM-HV.
>
> Summary for KVM-HV:
>
> -smp 1 -> works
> -smp 2 -> doesn't work
> -smp 3 -> works
> -smp 4 -> doesn't work
>
> I used -smp 4 before the RC1 of kernel 5.13 because my FSL P5040 BookE 
> machine [2] has 4 cores.
>
> Does this patch solve this issue? [3]
>
> Thanks,
> Christian
>
> [1] https://lists.ozlabs.org/pipermail/linuxppc-dev/2021-May/229103.html
> [2] http://wiki.amiga.org/index.php?title=X5000
> [3] 
> https://lists.ozlabs.org/pipermail/linuxppc-dev/2021-September/234152.html
Hello,

Since the RC5 of kernel 5.19, -smp 2 and -smp 4 work again. I don't know 
which patch has solved the issue.

Cheers,
Christian