[Qemu-devel] [RFC PATCH v3 0/5] s390: Host support for channel I/O.

[Qemu-devel] [RFC PATCH v3 0/5] s390: Host support for channel I/O.

[Cornelia Huck]

Hi,

here's the latest incarnation of my host patches to support channel
I/O on s390.

Most patches have only seen minor fixes, but patch 5 is completely
different since the kvm <-> user space interface has been reworked.

We now handle only interrupt-related operations in kvm. This
includes two channel I/O instructions that can dequeue pending I/O
interrupts: tpi and tsch (not the part actually interacting with
the subchannel). This makes the interface less complex (only one
new exit for tsch handling) and avoids duplicating code from qemu.

Cornelia Huck (5):
  KVM: s390: Support for I/O interrupts.
  KVM: s390: Add support for machine checks.
  KVM: s390: In-kernel handling of I/O instructions.
  KVM: s390: Base infrastructure for enabling capabilities.
  KVM: s390: Add support for channel I/O instructions.

 Documentation/virtual/kvm/api.txt |  40 +++++-
 arch/s390/include/asm/kvm_host.h  |  11 ++
 arch/s390/kvm/intercept.c         |  22 ++-
 arch/s390/kvm/interrupt.c         | 264 +++++++++++++++++++++++++++++++++++-
 arch/s390/kvm/kvm-s390.c          |  38 ++++++
 arch/s390/kvm/kvm-s390.h          |   6 +
 arch/s390/kvm/priv.c              | 275 +++++++++++++++++++++++++++++++++++---
 arch/s390/kvm/trace-s390.h        |  26 +++-
 include/linux/kvm.h               |  18 +++
 include/trace/events/kvm.h        |   2 +-
 10 files changed, 673 insertions(+), 29 deletions(-)

-- 
1.7.12.4

[Qemu-devel] [PATCH 1/5] KVM: s390: Support for I/O interrupts.

[Cornelia Huck]

Add support for handling I/O interrupts (standard, subchannel-related
ones and rudimentary adapter interrupts).

The subchannel-identifying parameters are encoded into the interrupt
type.

I/O interrupts are floating, so they can't be injected on a specific
vcpu.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 Documentation/virtual/kvm/api.txt |   4 ++
 arch/s390/include/asm/kvm_host.h  |   2 +
 arch/s390/kvm/interrupt.c         | 115 ++++++++++++++++++++++++++++++++++++--
 include/linux/kvm.h               |   6 ++
 4 files changed, 122 insertions(+), 5 deletions(-)

diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index 4258180..b660a6e 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -2068,6 +2068,10 @@ KVM_S390_INT_VIRTIO (vm) - virtio external interrupt; external interrupt
 KVM_S390_INT_SERVICE (vm) - sclp external interrupt; sclp parameter in parm
 KVM_S390_INT_EMERGENCY (vcpu) - sigp emergency; source cpu in parm
 KVM_S390_INT_EXTERNAL_CALL (vcpu) - sigp external call; source cpu in parm
+KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an
+    I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel);
+    I/O interruption parameters in parm (subchannel) and parm64 (intparm,
+    interruption subclass)
 
 Note that the vcpu ioctl is asynchronous to vcpu execution.
 
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index b784154..e47f697 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -76,6 +76,7 @@ struct kvm_s390_sie_block {
 	__u64	epoch;			/* 0x0038 */
 	__u8	reserved40[4];		/* 0x0040 */
 #define LCTL_CR0	0x8000
+#define LCTL_CR6	0x0200
 	__u16   lctl;			/* 0x0044 */
 	__s16	icpua;			/* 0x0046 */
 	__u32	ictl;			/* 0x0048 */
@@ -127,6 +128,7 @@ struct kvm_vcpu_stat {
 	u32 deliver_prefix_signal;
 	u32 deliver_restart_signal;
 	u32 deliver_program_int;
+	u32 deliver_io_int;
 	u32 exit_wait_state;
 	u32 instruction_stidp;
 	u32 instruction_spx;
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index c30615e..070ba22 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -21,11 +21,26 @@
 #include "gaccess.h"
 #include "trace-s390.h"
 
+#define IOINT_SCHID_MASK 0x0000ffff
+#define IOINT_SSID_MASK 0x00030000
+#define IOINT_CSSID_MASK 0x03fc0000
+#define IOINT_AI_MASK 0x04000000
+
+static int is_ioint(u64 type)
+{
+	return ((type & 0xfffe0000u) != 0xfffe0000u);
+}
+
 static int psw_extint_disabled(struct kvm_vcpu *vcpu)
 {
 	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_EXT);
 }
 
+static int psw_ioint_disabled(struct kvm_vcpu *vcpu)
+{
+	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_IO);
+}
+
 static int psw_interrupts_disabled(struct kvm_vcpu *vcpu)
 {
 	if ((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PER) ||
@@ -68,7 +83,18 @@ static int __interrupt_is_deliverable(struct kvm_vcpu *vcpu,
 	case KVM_S390_RESTART:
 		return 1;
 	default:
-		BUG();
+		if (is_ioint(inti->type)) {
+			if (psw_ioint_disabled(vcpu))
+				return 0;
+			if (vcpu->arch.sie_block->gcr[6] &
+			    inti->io.io_int_word)
+				return 1;
+			return 0;
+		} else {
+			printk(KERN_WARNING "illegal interrupt type %llx\n",
+			       inti->type);
+			BUG();
+		}
 	}
 	return 0;
 }
@@ -117,6 +143,13 @@ static void __set_intercept_indicator(struct kvm_vcpu *vcpu,
 		__set_cpuflag(vcpu, CPUSTAT_STOP_INT);
 		break;
 	default:
+		if (is_ioint(inti->type)) {
+			if (psw_ioint_disabled(vcpu))
+				__set_cpuflag(vcpu, CPUSTAT_IO_INT);
+			else
+				vcpu->arch.sie_block->lctl |= LCTL_CR6;
+			break;
+		}
 		BUG();
 	}
 }
@@ -298,7 +331,49 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu,
 		break;
 
 	default:
-		BUG();
+		if (is_ioint(inti->type)) {
+			__u32 param0 = ((__u32)inti->io.subchannel_id << 16) |
+				inti->io.subchannel_nr;
+			__u64 param1 = ((__u64)inti->io.io_int_parm << 32) |
+				inti->io.io_int_word;
+			VCPU_EVENT(vcpu, 4,
+				   "interrupt: I/O %llx", inti->type);
+			vcpu->stat.deliver_io_int++;
+			trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type,
+							 param0, param1);
+			rc = put_guest_u16(vcpu, __LC_SUBCHANNEL_ID,
+					   inti->io.subchannel_id);
+			if (rc == -EFAULT)
+				exception = 1;
+
+			rc = put_guest_u16(vcpu, __LC_SUBCHANNEL_NR,
+					   inti->io.subchannel_nr);
+			if (rc == -EFAULT)
+				exception = 1;
+
+			rc = put_guest_u32(vcpu, __LC_IO_INT_PARM,
+					   inti->io.io_int_parm);
+			if (rc == -EFAULT)
+				exception = 1;
+
+			rc = put_guest_u32(vcpu, __LC_IO_INT_WORD,
+					   inti->io.io_int_word);
+			if (rc == -EFAULT)
+				exception = 1;
+
+			rc = copy_to_guest(vcpu, __LC_IO_OLD_PSW,
+					   &vcpu->arch.sie_block->gpsw,
+					   sizeof(psw_t));
+			if (rc == -EFAULT)
+				exception = 1;
+
+			rc = copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw,
+					     __LC_IO_NEW_PSW, sizeof(psw_t));
+			if (rc == -EFAULT)
+				exception = 1;
+			break;
+		} else
+			BUG();
 	}
 	if (exception) {
 		printk("kvm: The guest lowcore is not mapped during interrupt "
@@ -545,7 +620,7 @@ int kvm_s390_inject_vm(struct kvm *kvm,
 {
 	struct kvm_s390_local_interrupt *li;
 	struct kvm_s390_float_interrupt *fi;
-	struct kvm_s390_interrupt_info *inti;
+	struct kvm_s390_interrupt_info *inti, *iter;
 	int sigcpu;
 
 	inti = kzalloc(sizeof(*inti), GFP_KERNEL);
@@ -569,9 +644,26 @@ int kvm_s390_inject_vm(struct kvm *kvm,
 	case KVM_S390_SIGP_STOP:
 	case KVM_S390_INT_EXTERNAL_CALL:
 	case KVM_S390_INT_EMERGENCY:
-	default:
 		kfree(inti);
 		return -EINVAL;
+	default:
+		if (!is_ioint(s390int->type)) {
+			kfree(inti);
+			return -EINVAL;
+		}
+		if (s390int->type & IOINT_AI_MASK)
+			VM_EVENT(kvm, 5, "%s", "inject: I/O (AI)");
+		else
+			VM_EVENT(kvm, 5, "inject: I/O css %x ss %x schid %04x",
+				 s390int->type & IOINT_CSSID_MASK,
+				 s390int->type & IOINT_SSID_MASK,
+				 s390int->type & IOINT_SCHID_MASK);
+		inti->type = s390int->type;
+		inti->io.subchannel_id = s390int->parm >> 16;
+		inti->io.subchannel_nr = s390int->parm & 0x0000ffffu;
+		inti->io.io_int_parm = s390int->parm64 >> 32;
+		inti->io.io_int_word = s390int->parm64 & 0x00000000ffffffffull;
+		break;
 	}
 	trace_kvm_s390_inject_vm(s390int->type, s390int->parm, s390int->parm64,
 				 2);
@@ -579,7 +671,19 @@ int kvm_s390_inject_vm(struct kvm *kvm,
 	mutex_lock(&kvm->lock);
 	fi = &kvm->arch.float_int;
 	spin_lock(&fi->lock);
-	list_add_tail(&inti->list, &fi->list);
+	if (!is_ioint(inti->type))
+		list_add_tail(&inti->list, &fi->list);
+	else {
+		/* Keep I/O interrupts sorted in isc order. */
+		list_for_each_entry(iter, &fi->list, list) {
+			if (!is_ioint(iter->type))
+				continue;
+			if (iter->io.io_int_word <= inti->io.io_int_word)
+				continue;
+			break;
+		}
+		list_add_tail(&inti->list, &iter->list);
+	}
 	atomic_set(&fi->active, 1);
 	sigcpu = find_first_bit(fi->idle_mask, KVM_MAX_VCPUS);
 	if (sigcpu == KVM_MAX_VCPUS) {
@@ -654,6 +758,7 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 	case KVM_S390_INT_VIRTIO:
 	case KVM_S390_INT_SERVICE:
 	default:
+		/* also includes IOINT */
 		kfree(inti);
 		return -EINVAL;
 	}
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 494a84c..e676940 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -401,6 +401,12 @@ struct kvm_s390_psw {
 #define KVM_S390_INT_SERVICE		0xffff2401u
 #define KVM_S390_INT_EMERGENCY		0xffff1201u
 #define KVM_S390_INT_EXTERNAL_CALL	0xffff1202u
+#define KVM_S390_INT_IO(ai,cssid,ssid,schid)   \
+	(((schid)) |			       \
+	 ((ssid) << 16) |		       \
+	 ((cssid) << 18) |		       \
+	 ((ai) << 26))
+
 
 struct kvm_s390_interrupt {
 	__u32 type;
-- 
1.7.12.4

[Qemu-devel] [PATCH 2/5] KVM: s390: Add support for machine checks.

[Cornelia Huck]

Add support for injecting machine checks (only repressible
conditions for now).

This is a bit more involved than I/O interrupts, for these reasons:

- Machine checks come in both floating and cpu varieties.
- We don't have a bit for machine checks enabling, but have to use
  a roundabout approach with trapping PSW changing instructions and
  watching for opened machine checks.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 Documentation/virtual/kvm/api.txt |   4 ++
 arch/s390/include/asm/kvm_host.h  |   8 +++
 arch/s390/kvm/intercept.c         |   2 +
 arch/s390/kvm/interrupt.c         | 112 ++++++++++++++++++++++++++++++++
 arch/s390/kvm/kvm-s390.h          |   3 +
 arch/s390/kvm/priv.c              | 133 ++++++++++++++++++++++++++++++++++++++
 arch/s390/kvm/trace-s390.h        |   6 +-
 include/linux/kvm.h               |   1 +
 8 files changed, 266 insertions(+), 3 deletions(-)

diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index b660a6e..5d09948 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -2072,6 +2072,10 @@ KVM_S390_INT_IO(ai,cssid,ssid,schid) (vm) - compound value to indicate an
     I/O interrupt (ai - adapter interrupt; cssid,ssid,schid - subchannel);
     I/O interruption parameters in parm (subchannel) and parm64 (intparm,
     interruption subclass)
+KVM_S390_MCHK (vm, vcpu) - machine check interrupt; cr 14 bits in parm,
+                           machine check interrupt code in parm64 (note that
+                           machine checks needing further payload are not
+                           supported by this ioctl)
 
 Note that the vcpu ioctl is asynchronous to vcpu execution.
 
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index e47f697..773859e 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -77,8 +77,10 @@ struct kvm_s390_sie_block {
 	__u8	reserved40[4];		/* 0x0040 */
 #define LCTL_CR0	0x8000
 #define LCTL_CR6	0x0200
+#define LCTL_CR14	0x0002
 	__u16   lctl;			/* 0x0044 */
 	__s16	icpua;			/* 0x0046 */
+#define ICTL_LPSW 0x00400000
 	__u32	ictl;			/* 0x0048 */
 	__u32	eca;			/* 0x004c */
 	__u8	icptcode;		/* 0x0050 */
@@ -189,6 +191,11 @@ struct kvm_s390_emerg_info {
 	__u16 code;
 };
 
+struct kvm_s390_mchk_info {
+	__u64 cr14;
+	__u64 mcic;
+};
+
 struct kvm_s390_interrupt_info {
 	struct list_head list;
 	u64	type;
@@ -199,6 +206,7 @@ struct kvm_s390_interrupt_info {
 		struct kvm_s390_emerg_info emerg;
 		struct kvm_s390_extcall_info extcall;
 		struct kvm_s390_prefix_info prefix;
+		struct kvm_s390_mchk_info mchk;
 	};
 };
 
diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index 22798ec..ec1177f 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -106,10 +106,12 @@ static int handle_lctl(struct kvm_vcpu *vcpu)
 
 static intercept_handler_t instruction_handlers[256] = {
 	[0x01] = kvm_s390_handle_01,
+	[0x82] = kvm_s390_handle_lpsw,
 	[0x83] = kvm_s390_handle_diag,
 	[0xae] = kvm_s390_handle_sigp,
 	[0xb2] = kvm_s390_handle_b2,
 	[0xb7] = handle_lctl,
+	[0xb9] = kvm_s390_handle_b9,
 	[0xe5] = kvm_s390_handle_e5,
 	[0xeb] = handle_lctlg,
 };
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index 070ba22..6b10267 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -41,6 +41,11 @@ static int psw_ioint_disabled(struct kvm_vcpu *vcpu)
 	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_IO);
 }
 
+static int psw_mchk_disabled(struct kvm_vcpu *vcpu)
+{
+	return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_MCHECK);
+}
+
 static int psw_interrupts_disabled(struct kvm_vcpu *vcpu)
 {
 	if ((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PER) ||
@@ -82,6 +87,12 @@ static int __interrupt_is_deliverable(struct kvm_vcpu *vcpu,
 	case KVM_S390_SIGP_SET_PREFIX:
 	case KVM_S390_RESTART:
 		return 1;
+	case KVM_S390_MCHK:
+		if (psw_mchk_disabled(vcpu))
+			return 0;
+		if (vcpu->arch.sie_block->gcr[14] & inti->mchk.cr14)
+			return 1;
+		return 0;
 	default:
 		if (is_ioint(inti->type)) {
 			if (psw_ioint_disabled(vcpu))
@@ -119,6 +130,7 @@ static void __reset_intercept_indicators(struct kvm_vcpu *vcpu)
 		CPUSTAT_IO_INT | CPUSTAT_EXT_INT | CPUSTAT_STOP_INT,
 		&vcpu->arch.sie_block->cpuflags);
 	vcpu->arch.sie_block->lctl = 0x0000;
+	vcpu->arch.sie_block->ictl &= ~ICTL_LPSW;
 }
 
 static void __set_cpuflag(struct kvm_vcpu *vcpu, u32 flag)
@@ -142,6 +154,12 @@ static void __set_intercept_indicator(struct kvm_vcpu *vcpu,
 	case KVM_S390_SIGP_STOP:
 		__set_cpuflag(vcpu, CPUSTAT_STOP_INT);
 		break;
+	case KVM_S390_MCHK:
+		if (psw_mchk_disabled(vcpu))
+			vcpu->arch.sie_block->ictl |= ICTL_LPSW;
+		else
+			vcpu->arch.sie_block->lctl |= LCTL_CR14;
+		break;
 	default:
 		if (is_ioint(inti->type)) {
 			if (psw_ioint_disabled(vcpu))
@@ -330,6 +348,32 @@ static void __do_deliver_interrupt(struct kvm_vcpu *vcpu,
 			exception = 1;
 		break;
 
+	case KVM_S390_MCHK:
+		VCPU_EVENT(vcpu, 4, "interrupt: machine check mcic=%llx",
+			   inti->mchk.mcic);
+		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id, inti->type,
+						 inti->mchk.cr14,
+						 inti->mchk.mcic);
+		rc = kvm_s390_vcpu_store_status(vcpu,
+						KVM_S390_STORE_STATUS_PREFIXED);
+		if (rc == -EFAULT)
+			exception = 1;
+
+		rc = put_guest_u64(vcpu, __LC_MCCK_CODE, inti->mchk.mcic);
+		if (rc == -EFAULT)
+			exception = 1;
+
+		rc = copy_to_guest(vcpu, __LC_MCK_OLD_PSW,
+				   &vcpu->arch.sie_block->gpsw, sizeof(psw_t));
+		if (rc == -EFAULT)
+			exception = 1;
+
+		rc = copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw,
+				     __LC_MCK_NEW_PSW, sizeof(psw_t));
+		if (rc == -EFAULT)
+			exception = 1;
+		break;
+
 	default:
 		if (is_ioint(inti->type)) {
 			__u32 param0 = ((__u32)inti->io.subchannel_id << 16) |
@@ -593,6 +637,61 @@ void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
 	}
 }
 
+void kvm_s390_deliver_pending_machine_checks(struct kvm_vcpu *vcpu)
+{
+	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
+	struct kvm_s390_float_interrupt *fi = vcpu->arch.local_int.float_int;
+	struct kvm_s390_interrupt_info  *n, *inti = NULL;
+	int deliver;
+
+	__reset_intercept_indicators(vcpu);
+	if (atomic_read(&li->active)) {
+		do {
+			deliver = 0;
+			spin_lock_bh(&li->lock);
+			list_for_each_entry_safe(inti, n, &li->list, list) {
+				if ((inti->type == KVM_S390_MCHK) &&
+				    __interrupt_is_deliverable(vcpu, inti)) {
+					list_del(&inti->list);
+					deliver = 1;
+					break;
+				}
+				__set_intercept_indicator(vcpu, inti);
+			}
+			if (list_empty(&li->list))
+				atomic_set(&li->active, 0);
+			spin_unlock_bh(&li->lock);
+			if (deliver) {
+				__do_deliver_interrupt(vcpu, inti);
+				kfree(inti);
+			}
+		} while (deliver);
+	}
+
+	if (atomic_read(&fi->active)) {
+		do {
+			deliver = 0;
+			spin_lock(&fi->lock);
+			list_for_each_entry_safe(inti, n, &fi->list, list) {
+				if ((inti->type == KVM_S390_MCHK) &&
+				    __interrupt_is_deliverable(vcpu, inti)) {
+					list_del(&inti->list);
+					deliver = 1;
+					break;
+				}
+				__set_intercept_indicator(vcpu, inti);
+			}
+			if (list_empty(&fi->list))
+				atomic_set(&fi->active, 0);
+			spin_unlock(&fi->lock);
+			if (deliver) {
+				__do_deliver_interrupt(vcpu, inti);
+				kfree(inti);
+			}
+		} while (deliver);
+	}
+}
+
 int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code)
 {
 	struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int;
@@ -646,6 +745,13 @@ int kvm_s390_inject_vm(struct kvm *kvm,
 	case KVM_S390_INT_EMERGENCY:
 		kfree(inti);
 		return -EINVAL;
+	case KVM_S390_MCHK:
+		VM_EVENT(kvm, 5, "inject: machine check parm64:%llx",
+			 s390int->parm64);
+		inti->type = s390int->type;
+		inti->mchk.cr14 = s390int->parm; /* upper bits are not used */
+		inti->mchk.mcic = s390int->parm64;
+		break;
 	default:
 		if (!is_ioint(s390int->type)) {
 			kfree(inti);
@@ -755,6 +861,12 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 		inti->type = s390int->type;
 		inti->emerg.code = s390int->parm;
 		break;
+	case KVM_S390_MCHK:
+		VCPU_EVENT(vcpu, 5, "inject: machine check parm64:%llx",
+			   s390int->parm64);
+		inti->type = s390int->type;
+		inti->mchk.mcic = s390int->parm64;
+		break;
 	case KVM_S390_INT_VIRTIO:
 	case KVM_S390_INT_SERVICE:
 	default:
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index d75bc5e..b1e1cb6 100644
--- a/arch/s390/kvm/kvm-s390.h
+++ b/arch/s390/kvm/kvm-s390.h
@@ -69,6 +69,7 @@ int kvm_s390_handle_wait(struct kvm_vcpu *vcpu);
 enum hrtimer_restart kvm_s390_idle_wakeup(struct hrtimer *timer);
 void kvm_s390_tasklet(unsigned long parm);
 void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu);
+void kvm_s390_deliver_pending_machine_checks(struct kvm_vcpu *vcpu);
 int kvm_s390_inject_vm(struct kvm *kvm,
 		struct kvm_s390_interrupt *s390int);
 int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
@@ -80,6 +81,8 @@ int kvm_s390_inject_sigp_stop(struct kvm_vcpu *vcpu, int action);
 int kvm_s390_handle_b2(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_e5(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_01(struct kvm_vcpu *vcpu);
+int kvm_s390_handle_b9(struct kvm_vcpu *vcpu);
+int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu);
 
 /* implemented in sigp.c */
 int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu);
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index d768906..bf13ce9 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -176,6 +176,101 @@ static int handle_stfl(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
+static void handle_new_psw(struct kvm_vcpu *vcpu)
+{
+	/* Check whether the new psw is enabled for machine checks. */
+	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_MCHECK)
+		kvm_s390_deliver_pending_machine_checks(vcpu);
+}
+
+int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu)
+{
+	int base2 = vcpu->arch.sie_block->ipb >> 28;
+	int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16);
+	u64 addr;
+	u64 uninitialized_var(new_psw);
+
+	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
+		return kvm_s390_inject_program_int(vcpu,
+						   PGM_PRIVILEGED_OPERATION);
+
+	addr = disp2;
+	if (base2)
+		addr += vcpu->run->s.regs.gprs[base2];
+
+	if (addr & 7) {
+		kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+		goto out;
+	}
+
+	if (get_guest_u64(vcpu, addr, &new_psw)) {
+		kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+		goto out;
+	}
+
+	if (!(new_psw & 0x0008000000000000)) {
+		kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+		goto out;
+	}
+
+	vcpu->arch.sie_block->gpsw.mask = new_psw & 0xfff7ffff80000000;
+	vcpu->arch.sie_block->gpsw.addr = new_psw & 0x000000007fffffff;
+
+	if ((vcpu->arch.sie_block->gpsw.mask & 0xb80800fe00000000) ||
+	    (!(vcpu->arch.sie_block->gpsw.mask & 0x0000000180000000) &&
+	     (vcpu->arch.sie_block->gpsw.addr & 0x000000007ff00000)) ||
+	    ((vcpu->arch.sie_block->gpsw.mask & 0x0000000110000000) ==
+	     0x0000000100000000)) {
+		kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+		goto out;
+	}
+
+	handle_new_psw(vcpu);
+out:
+	return 0;
+}
+
+static int handle_lpswe(struct kvm_vcpu *vcpu)
+{
+	int base2 = vcpu->arch.sie_block->ipb >> 28;
+	int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16);
+	u64 addr;
+	psw_t new_psw;
+
+	addr = disp2;
+	if (base2)
+		addr += vcpu->run->s.regs.gprs[base2];
+
+	if (addr & 7) {
+		kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+		goto out;
+	}
+
+	if (copy_from_guest(vcpu, &new_psw, addr, sizeof(new_psw))) {
+		kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+		goto out;
+	}
+
+	vcpu->arch.sie_block->gpsw.mask = new_psw.mask;
+	vcpu->arch.sie_block->gpsw.addr = new_psw.addr;
+
+	if ((vcpu->arch.sie_block->gpsw.mask & 0xb80800fe7fffffff) ||
+	    (((vcpu->arch.sie_block->gpsw.mask & 0x0000000110000000) ==
+	      0x0000000010000000) &&
+	     (vcpu->arch.sie_block->gpsw.addr & 0xffffffff80000000)) ||
+	    (!(vcpu->arch.sie_block->gpsw.mask & 0x0000000180000000) &&
+	     (vcpu->arch.sie_block->gpsw.addr & 0xfffffffffff00000)) ||
+	    ((vcpu->arch.sie_block->gpsw.mask & 0x0000000110000000) ==
+	     0x0000000100000000)) {
+		kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+		goto out;
+	}
+
+	handle_new_psw(vcpu);
+out:
+	return 0;
+}
+
 static int handle_stidp(struct kvm_vcpu *vcpu)
 {
 	int base2 = vcpu->arch.sie_block->ipb >> 28;
@@ -309,6 +404,7 @@ static intercept_handler_t priv_handlers[256] = {
 	[0x5f] = handle_chsc,
 	[0x7d] = handle_stsi,
 	[0xb1] = handle_stfl,
+	[0xb2] = handle_lpswe,
 };
 
 int kvm_s390_handle_b2(struct kvm_vcpu *vcpu)
@@ -333,6 +429,43 @@ int kvm_s390_handle_b2(struct kvm_vcpu *vcpu)
 	return -EOPNOTSUPP;
 }
 
+static int handle_epsw(struct kvm_vcpu *vcpu)
+{
+	int reg1, reg2;
+
+	reg1 = (vcpu->arch.sie_block->ipb & 0x00f00000) >> 24;
+	reg2 = (vcpu->arch.sie_block->ipb & 0x000f0000) >> 16;
+	vcpu->run->s.regs.gprs[reg1] &= 0xffffffff00000000;
+	vcpu->run->s.regs.gprs[reg1] |= vcpu->arch.sie_block->gpsw.mask >> 32;
+	if (reg2) {
+		vcpu->run->s.regs.gprs[reg2] &= 0xffffffff00000000;
+		vcpu->run->s.regs.gprs[reg2] |=
+			vcpu->arch.sie_block->gpsw.mask & 0x00000000ffffffff;
+	}
+	return 0;
+}
+
+static intercept_handler_t b9_handlers[256] = {
+	[0x8d] = handle_epsw,
+};
+
+int kvm_s390_handle_b9(struct kvm_vcpu *vcpu)
+{
+	intercept_handler_t handler;
+
+	/* This is handled just as for the B2 instructions. */
+	handler = b9_handlers[vcpu->arch.sie_block->ipa & 0x00ff];
+	if (handler) {
+		if ((handler != handle_epsw) &&
+		    (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE))
+			return kvm_s390_inject_program_int(vcpu,
+						   PGM_PRIVILEGED_OPERATION);
+		else
+			return handler(vcpu);
+	}
+	return -EOPNOTSUPP;
+}
+
 static int handle_tprot(struct kvm_vcpu *vcpu)
 {
 	int base1 = (vcpu->arch.sie_block->ipb & 0xf0000000) >> 28;
diff --git a/arch/s390/kvm/trace-s390.h b/arch/s390/kvm/trace-s390.h
index 90fdf85..95fbc1a 100644
--- a/arch/s390/kvm/trace-s390.h
+++ b/arch/s390/kvm/trace-s390.h
@@ -141,13 +141,13 @@ TRACE_EVENT(kvm_s390_inject_vcpu,
  * Trace point for the actual delivery of interrupts.
  */
 TRACE_EVENT(kvm_s390_deliver_interrupt,
-	    TP_PROTO(unsigned int id, __u64 type, __u32 data0, __u64 data1),
+	    TP_PROTO(unsigned int id, __u64 type, __u64 data0, __u64 data1),
 	    TP_ARGS(id, type, data0, data1),
 
 	    TP_STRUCT__entry(
 		    __field(int, id)
 		    __field(__u32, inttype)
-		    __field(__u32, data0)
+		    __field(__u64, data0)
 		    __field(__u64, data1)
 		    ),
 
@@ -159,7 +159,7 @@ TRACE_EVENT(kvm_s390_deliver_interrupt,
 		    ),
 
 	    TP_printk("deliver interrupt (vcpu %d): type:%x (%s) "	\
-		      "data:%08x %016llx",
+		      "data:%08llx %016llx",
 		      __entry->id, __entry->inttype,
 		      __print_symbolic(__entry->inttype, kvm_s390_int_type),
 		      __entry->data0, __entry->data1)
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index e676940..22859dc 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -397,6 +397,7 @@ struct kvm_s390_psw {
 #define KVM_S390_PROGRAM_INT		0xfffe0001u
 #define KVM_S390_SIGP_SET_PREFIX	0xfffe0002u
 #define KVM_S390_RESTART		0xfffe0003u
+#define KVM_S390_MCHK			0xfffe1000u
 #define KVM_S390_INT_VIRTIO		0xffff2603u
 #define KVM_S390_INT_SERVICE		0xffff2401u
 #define KVM_S390_INT_EMERGENCY		0xffff1201u
-- 
1.7.12.4

[Qemu-devel] [PATCH 3/5] KVM: s390: In-kernel handling of I/O instructions.

[Cornelia Huck]

Explicitely catch all channel I/O related instructions intercepts
in the kernel and set condition code 3 for them.

This paws the way for properly handling these instructions later
on.

Note: This is not architecture compliant (the previous code wasn't
either) since setting cc 3 is not the correct thing to do for some
of these instructions. For Linux guests, however, it still has the
intended effect of stopping css probing.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 arch/s390/kvm/intercept.c | 19 +++++++++++++---
 arch/s390/kvm/kvm-s390.h  |  1 +
 arch/s390/kvm/priv.c      | 56 +++++++++++++++++++++++++++++++++--------------
 3 files changed, 56 insertions(+), 20 deletions(-)

diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index ec1177f..754dc9e 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -33,8 +33,6 @@ static int handle_lctlg(struct kvm_vcpu *vcpu)
 	int reg, rc;
 
 	vcpu->stat.instruction_lctlg++;
-	if ((vcpu->arch.sie_block->ipb & 0xff) != 0x2f)
-		return -EOPNOTSUPP;
 
 	useraddr = disp2;
 	if (base2)
@@ -104,6 +102,21 @@ static int handle_lctl(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
+static intercept_handler_t eb_handlers[256] = {
+	[0x2f] = handle_lctlg,
+	[0x8a] = kvm_s390_handle_priv_eb,
+};
+
+static int handle_eb(struct kvm_vcpu *vcpu)
+{
+	intercept_handler_t handler;
+
+	handler = eb_handlers[vcpu->arch.sie_block->ipb & 0xff];
+	if (handler)
+		return handler(vcpu);
+	return -EOPNOTSUPP;
+}
+
 static intercept_handler_t instruction_handlers[256] = {
 	[0x01] = kvm_s390_handle_01,
 	[0x82] = kvm_s390_handle_lpsw,
@@ -113,7 +126,7 @@ static intercept_handler_t instruction_handlers[256] = {
 	[0xb7] = handle_lctl,
 	[0xb9] = kvm_s390_handle_b9,
 	[0xe5] = kvm_s390_handle_e5,
-	[0xeb] = handle_lctlg,
+	[0xeb] = handle_eb,
 };
 
 static int handle_noop(struct kvm_vcpu *vcpu)
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index b1e1cb6..7f50229 100644
--- a/arch/s390/kvm/kvm-s390.h
+++ b/arch/s390/kvm/kvm-s390.h
@@ -83,6 +83,7 @@ int kvm_s390_handle_e5(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_01(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_b9(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu);
+int kvm_s390_handle_priv_eb(struct kvm_vcpu *vcpu);
 
 /* implemented in sigp.c */
 int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu);
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index bf13ce9..2aba96b 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -135,20 +135,9 @@ static int handle_skey(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-static int handle_stsch(struct kvm_vcpu *vcpu)
+static int handle_io_inst(struct kvm_vcpu *vcpu)
 {
-	vcpu->stat.instruction_stsch++;
-	VCPU_EVENT(vcpu, 4, "%s", "store subchannel - CC3");
-	/* condition code 3 */
-	vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44);
-	vcpu->arch.sie_block->gpsw.mask |= (3 & 3ul) << 44;
-	return 0;
-}
-
-static int handle_chsc(struct kvm_vcpu *vcpu)
-{
-	vcpu->stat.instruction_chsc++;
-	VCPU_EVENT(vcpu, 4, "%s", "channel subsystem call - CC3");
+	VCPU_EVENT(vcpu, 4, "%s", "I/O instruction");
 	/* condition code 3 */
 	vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44);
 	vcpu->arch.sie_block->gpsw.mask |= (3 & 3ul) << 44;
@@ -392,7 +381,7 @@ out_fail:
 	return 0;
 }
 
-static intercept_handler_t priv_handlers[256] = {
+static intercept_handler_t b2_handlers[256] = {
 	[0x02] = handle_stidp,
 	[0x10] = handle_set_prefix,
 	[0x11] = handle_store_prefix,
@@ -400,8 +389,22 @@ static intercept_handler_t priv_handlers[256] = {
 	[0x29] = handle_skey,
 	[0x2a] = handle_skey,
 	[0x2b] = handle_skey,
-	[0x34] = handle_stsch,
-	[0x5f] = handle_chsc,
+	[0x30] = handle_io_inst,
+	[0x31] = handle_io_inst,
+	[0x32] = handle_io_inst,
+	[0x33] = handle_io_inst,
+	[0x34] = handle_io_inst,
+	[0x35] = handle_io_inst,
+	[0x36] = handle_io_inst,
+	[0x37] = handle_io_inst,
+	[0x38] = handle_io_inst,
+	[0x39] = handle_io_inst,
+	[0x3a] = handle_io_inst,
+	[0x3b] = handle_io_inst,
+	[0x3c] = handle_io_inst,
+	[0x5f] = handle_io_inst,
+	[0x74] = handle_io_inst,
+	[0x76] = handle_io_inst,
 	[0x7d] = handle_stsi,
 	[0xb1] = handle_stfl,
 	[0xb2] = handle_lpswe,
@@ -418,7 +421,7 @@ int kvm_s390_handle_b2(struct kvm_vcpu *vcpu)
 	 * state bit and (a) handle the instruction or (b) send a code 2
 	 * program check.
 	 * Anything else goes to userspace.*/
-	handler = priv_handlers[vcpu->arch.sie_block->ipa & 0x00ff];
+	handler = b2_handlers[vcpu->arch.sie_block->ipa & 0x00ff];
 	if (handler) {
 		if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
 			return kvm_s390_inject_program_int(vcpu,
@@ -447,6 +450,7 @@ static int handle_epsw(struct kvm_vcpu *vcpu)
 
 static intercept_handler_t b9_handlers[256] = {
 	[0x8d] = handle_epsw,
+	[0x9c] = handle_io_inst,
 };
 
 int kvm_s390_handle_b9(struct kvm_vcpu *vcpu)
@@ -466,6 +470,24 @@ int kvm_s390_handle_b9(struct kvm_vcpu *vcpu)
 	return -EOPNOTSUPP;
 }
 
+static intercept_handler_t eb_handlers[256] = {
+	[0x8a] = handle_io_inst,
+};
+
+int kvm_s390_handle_priv_eb(struct kvm_vcpu *vcpu)
+{
+	intercept_handler_t handler;
+
+	/* All eb instructions that end up here are privileged. */
+	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
+		return kvm_s390_inject_program_int(vcpu,
+						   PGM_PRIVILEGED_OPERATION);
+	handler = eb_handlers[vcpu->arch.sie_block->ipb & 0xff];
+	if (handler)
+		return handler(vcpu);
+	return -EOPNOTSUPP;
+}
+
 static int handle_tprot(struct kvm_vcpu *vcpu)
 {
 	int base1 = (vcpu->arch.sie_block->ipb & 0xf0000000) >> 28;
-- 
1.7.12.4

[Qemu-devel] [PATCH 4/5] KVM: s390: Base infrastructure for enabling capabilities.

[Cornelia Huck]

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 Documentation/virtual/kvm/api.txt |  2 +-
 arch/s390/kvm/kvm-s390.c          | 26 ++++++++++++++++++++++++++
 2 files changed, 27 insertions(+), 1 deletion(-)

diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index 5d09948..a0408cf 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -913,7 +913,7 @@ documentation when it pops into existence).
 4.37 KVM_ENABLE_CAP
 
 Capability: KVM_CAP_ENABLE_CAP
-Architectures: ppc
+Architectures: ppc, s390
 Type: vcpu ioctl
 Parameters: struct kvm_enable_cap (in)
 Returns: 0 on success; -1 on error
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 38883f0..c19568b 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -140,6 +140,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 #endif
 	case KVM_CAP_SYNC_REGS:
 	case KVM_CAP_ONE_REG:
+	case KVM_CAP_ENABLE_CAP:
 		r = 1;
 		break;
 	case KVM_CAP_NR_VCPUS:
@@ -807,6 +808,22 @@ int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr)
 	return 0;
 }
 
+static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu,
+				     struct kvm_enable_cap *cap)
+{
+	int r;
+
+	if (cap->flags)
+		return -EINVAL;
+
+	switch (cap->cap) {
+	default:
+		r = -EINVAL;
+		break;
+	}
+	return r;
+}
+
 long kvm_arch_vcpu_ioctl(struct file *filp,
 			 unsigned int ioctl, unsigned long arg)
 {
@@ -893,6 +910,15 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
 			r = 0;
 		break;
 	}
+	case KVM_ENABLE_CAP:
+	{
+		struct kvm_enable_cap cap;
+		r = -EFAULT;
+		if (copy_from_user(&cap, argp, sizeof(cap)))
+			break;
+		r = kvm_vcpu_ioctl_enable_cap(vcpu, &cap);
+		break;
+	}
 	default:
 		r = -ENOTTY;
 	}
-- 
1.7.12.4

[Qemu-devel] [PATCH 5/5] KVM: s390: Add support for channel I/O instructions.

[Cornelia Huck]

Add a new capability, KVM_CAP_S390_CSS_SUPPORT, which will pass
intercepts for channel I/O instructions to userspace. Only I/O
instructions interacting with I/O interrupts need to be handled
in-kernel:

- TEST PENDING INTERRUPTION (tpi) dequeues and stores pending
  interrupts entirely in-kernel.
- TEST SUBCHANNEL (tsch) dequeues pending interrupts in-kernel
  and exits via KVM_EXIT_S390_TSCH to userspace for subchannel-
  related processing.

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
---
 Documentation/virtual/kvm/api.txt | 30 ++++++++++++
 arch/s390/include/asm/kvm_host.h  |  1 +
 arch/s390/kvm/intercept.c         |  1 +
 arch/s390/kvm/interrupt.c         | 37 +++++++++++++++
 arch/s390/kvm/kvm-s390.c          | 12 +++++
 arch/s390/kvm/kvm-s390.h          |  2 +
 arch/s390/kvm/priv.c              | 96 +++++++++++++++++++++++++++++++++++++--
 arch/s390/kvm/trace-s390.h        | 20 ++++++++
 include/linux/kvm.h               | 11 +++++
 include/trace/events/kvm.h        |  2 +-
 10 files changed, 207 insertions(+), 5 deletions(-)

diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index a0408cf..2233081 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -2294,6 +2294,22 @@ The possible hypercalls are defined in the Power Architecture Platform
 Requirements (PAPR) document available from www.power.org (free
 developer registration required to access it).
 
+		/* KVM_EXIT_S390_TSCH */
+		struct {
+			__u16 subchannel_id;
+			__u16 subchannel_nr;
+			__u32 io_int_parm;
+			__u32 io_int_word;
+			__u32 ipb;
+			__u8 dequeued;
+		} s390_tsch;
+
+s390 specific. This exit occurs when KVM_CAP_S390_CSS_SUPPORT has been enabled
+and TEST SUBCHANNEL was intercepted. If dequeued is set, a pending I/O
+interrupt for the target subchannel has been dequeued and subchannel_id,
+subchannel_nr, io_int_parm and io_int_word contain the parameters for that
+interrupt. ipb is needed for instruction parameter decoding.
+
 		/* Fix the size of the union. */
 		char padding[256];
 	};
@@ -2415,3 +2431,17 @@ For mmu types KVM_MMU_FSL_BOOKE_NOHV and KVM_MMU_FSL_BOOKE_HV:
    where "num_sets" is the tlb_sizes[] value divided by the tlb_ways[] value.
  - The tsize field of mas1 shall be set to 4K on TLB0, even though the
    hardware ignores this value for TLB0.
+
+6.4 KVM_CAP_S390_CSS_SUPPORT
+
+Architectures: s390
+Parameters: none
+Returns: 0 on success; -1 on error
+
+This capability enables support for handling of channel I/O instructions.
+
+TEST PENDING INTERRUPTION and the interrupt portion of TEST SUBCHANNEL are
+handled in-kernel, while the other I/O instructions are passed to userspace.
+
+When this capability is enabled, KVM_EXIT_S390_TSCH will occur on TEST
+SUBCHANNEL intercepts.
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 773859e..091c581 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -264,6 +264,7 @@ struct kvm_arch{
 	debug_info_t *dbf;
 	struct kvm_s390_float_interrupt float_int;
 	struct gmap *gmap;
+	int css_support;
 };
 
 extern int sie64a(struct kvm_s390_sie_block *, u64 *);
diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index 754dc9e..9ab2efd 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -273,6 +273,7 @@ static const intercept_handler_t intercept_funcs[] = {
 	[0x0C >> 2] = handle_instruction_and_prog,
 	[0x10 >> 2] = handle_noop,
 	[0x14 >> 2] = handle_noop,
+	[0x18 >> 2] = handle_noop,
 	[0x1C >> 2] = kvm_s390_handle_wait,
 	[0x20 >> 2] = handle_validity,
 	[0x28 >> 2] = handle_stop,
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index 6b10267..495a411 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -714,6 +714,43 @@ int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code)
 	return 0;
 }
 
+struct kvm_s390_interrupt_info *kvm_s390_get_io_int(struct kvm *kvm,
+						    u64 cr6, u64 schid)
+{
+	struct kvm_s390_float_interrupt *fi;
+	struct kvm_s390_interrupt_info *inti, *iter;
+
+	if ((!schid && !cr6) || (schid && cr6))
+		return NULL;
+	mutex_lock(&kvm->lock);
+	fi = &kvm->arch.float_int;
+	spin_lock(&fi->lock);
+	inti = NULL;
+	list_for_each_entry(iter, &fi->list, list) {
+		if (!is_ioint(iter->type))
+			continue;
+		if (cr6 && ((cr6 & iter->io.io_int_word) == 0))
+			continue;
+		if (schid) {
+			if (((schid & 0x00000000ffff0000) >> 16) !=
+			    iter->io.subchannel_id)
+				continue;
+			if ((schid & 0x000000000000ffff) !=
+			    iter->io.subchannel_nr)
+				continue;
+		}
+		inti = iter;
+		break;
+	}
+	if (inti)
+		list_del_init(&inti->list);
+	if (list_empty(&fi->list))
+		atomic_set(&fi->active, 0);
+	spin_unlock(&fi->lock);
+	mutex_unlock(&kvm->lock);
+	return inti;
+}
+
 int kvm_s390_inject_vm(struct kvm *kvm,
 		       struct kvm_s390_interrupt *s390int)
 {
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index c19568b..fcbb711 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -141,6 +141,7 @@ int kvm_dev_ioctl_check_extension(long ext)
 	case KVM_CAP_SYNC_REGS:
 	case KVM_CAP_ONE_REG:
 	case KVM_CAP_ENABLE_CAP:
+	case KVM_CAP_S390_CSS_SUPPORT:
 		r = 1;
 		break;
 	case KVM_CAP_NR_VCPUS:
@@ -235,6 +236,9 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 		if (!kvm->arch.gmap)
 			goto out_nogmap;
 	}
+
+	kvm->arch.css_support = 0;
+
 	return 0;
 out_nogmap:
 	debug_unregister(kvm->arch.dbf);
@@ -657,6 +661,7 @@ rerun_vcpu:
 	case KVM_EXIT_INTR:
 	case KVM_EXIT_S390_RESET:
 	case KVM_EXIT_S390_UCONTROL:
+	case KVM_EXIT_S390_TSCH:
 		break;
 	default:
 		BUG();
@@ -817,6 +822,13 @@ static int kvm_vcpu_ioctl_enable_cap(struct kvm_vcpu *vcpu,
 		return -EINVAL;
 
 	switch (cap->cap) {
+	case KVM_CAP_S390_CSS_SUPPORT:
+		if (!vcpu->kvm->arch.css_support) {
+			vcpu->kvm->arch.css_support = 1;
+			trace_kvm_s390_enable_css(vcpu->kvm);
+		}
+		r = 0;
+		break;
 	default:
 		r = -EINVAL;
 		break;
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index 7f50229..14a417f 100644
--- a/arch/s390/kvm/kvm-s390.h
+++ b/arch/s390/kvm/kvm-s390.h
@@ -76,6 +76,8 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 		struct kvm_s390_interrupt *s390int);
 int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code);
 int kvm_s390_inject_sigp_stop(struct kvm_vcpu *vcpu, int action);
+struct kvm_s390_interrupt_info *kvm_s390_get_io_int(struct kvm *kvm,
+						    u64 cr6, u64 schid);
 
 /* implemented in priv.c */
 int kvm_s390_handle_b2(struct kvm_vcpu *vcpu);
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index 2aba96b..7473888 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -135,15 +135,103 @@ static int handle_skey(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-static int handle_io_inst(struct kvm_vcpu *vcpu)
+static int handle_tpi(struct kvm_vcpu *vcpu)
 {
-	VCPU_EVENT(vcpu, 4, "%s", "I/O instruction");
-	/* condition code 3 */
+	u32 ipb = vcpu->arch.sie_block->ipb;
+	u64 addr;
+	struct kvm_s390_interrupt_info *inti;
+	int cc;
+
+	addr = ipb >> 28;
+	if (addr > 0)
+		addr = vcpu->run->s.regs.gprs[addr];
+
+	addr += (ipb & 0xfff0000) >> 16;
+
+	inti = kvm_s390_get_io_int(vcpu->kvm, vcpu->run->s.regs.crs[6], 0);
+	if (inti) {
+		if (addr) {
+			/*
+			 * Store the two-word I/O interruption code into the
+			 * provided area.
+			 */
+			put_guest_u16(vcpu, addr, inti->io.subchannel_id);
+			put_guest_u16(vcpu, addr + 2, inti->io.subchannel_nr);
+			put_guest_u32(vcpu, addr + 4, inti->io.io_int_parm);
+		} else {
+			/*
+			 * Store the three-word I/O interruption code into
+			 * the appropriate lowcore area.
+			 */
+			put_guest_u16(vcpu, 184, inti->io.subchannel_id);
+			put_guest_u16(vcpu, 186, inti->io.subchannel_nr);
+			put_guest_u32(vcpu, 188, inti->io.io_int_parm);
+			put_guest_u32(vcpu, 192, inti->io.io_int_word);
+		}
+		cc = 1;
+	} else
+		cc = 0;
+	kfree(inti);
+	/* Set condition code and we're done. */
 	vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44);
-	vcpu->arch.sie_block->gpsw.mask |= (3 & 3ul) << 44;
+	vcpu->arch.sie_block->gpsw.mask |= (cc & 3ul) << 44;
 	return 0;
 }
 
+static int handle_tsch(struct kvm_vcpu *vcpu)
+{
+	struct kvm_s390_interrupt_info *inti;
+
+	inti = kvm_s390_get_io_int(vcpu->kvm, 0,
+				   vcpu->run->s.regs.gprs[1]);
+
+	/*
+	 * Prepare exit to userspace.
+	 * We indicate whether we dequeued a pending I/O interrupt
+	 * so that userspace can re-inject it if the instruction gets
+	 * a program check. While this may re-order the pending I/O
+	 * interrupts, this is no problem since the priority is kept
+	 * intact.
+	 */
+	vcpu->run->exit_reason = KVM_EXIT_S390_TSCH;
+	vcpu->run->s390_tsch.dequeued = !!inti;
+	if (inti) {
+		vcpu->run->s390_tsch.subchannel_id = inti->io.subchannel_id;
+		vcpu->run->s390_tsch.subchannel_nr = inti->io.subchannel_nr;
+		vcpu->run->s390_tsch.io_int_parm = inti->io.io_int_parm;
+		vcpu->run->s390_tsch.io_int_word = inti->io.io_int_word;
+	}
+	vcpu->run->s390_tsch.ipb = vcpu->arch.sie_block->ipb;
+	kfree(inti);
+	return -EREMOTE;
+}
+
+static int handle_io_inst(struct kvm_vcpu *vcpu)
+{
+	VCPU_EVENT(vcpu, 4, "%s", "I/O instruction");
+
+	if (vcpu->kvm->arch.css_support) {
+		/*
+		 * Most I/O instructions will be handled by userspace.
+		 * Exceptions are tpi and the interrupt portion of tsch.
+		 */
+		if (vcpu->arch.sie_block->ipa == 0xb236)
+			return handle_tpi(vcpu);
+		if (vcpu->arch.sie_block->ipa == 0xb235)
+			return handle_tsch(vcpu);
+		/* Handle in userspace. */
+		return -EOPNOTSUPP;
+	} else {
+		/*
+		 * Set condition code 3 to stop the guest from issueing channel
+		 * I/O instructions.
+		 */
+		vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44);
+		vcpu->arch.sie_block->gpsw.mask |= (3 & 3ul) << 44;
+		return 0;
+	}
+}
+
 static int handle_stfl(struct kvm_vcpu *vcpu)
 {
 	unsigned int facility_list;
diff --git a/arch/s390/kvm/trace-s390.h b/arch/s390/kvm/trace-s390.h
index 95fbc1a..13f30f5 100644
--- a/arch/s390/kvm/trace-s390.h
+++ b/arch/s390/kvm/trace-s390.h
@@ -204,6 +204,26 @@ TRACE_EVENT(kvm_s390_stop_request,
 	);
 
 
+/*
+ * Trace point for enabling channel I/O instruction support.
+ */
+TRACE_EVENT(kvm_s390_enable_css,
+	    TP_PROTO(void *kvm),
+	    TP_ARGS(kvm),
+
+	    TP_STRUCT__entry(
+		    __field(void *, kvm)
+		    ),
+
+	    TP_fast_assign(
+		    __entry->kvm = kvm;
+		    ),
+
+	    TP_printk("enabling channel I/O support (kvm @ %p)\n",
+		      __entry->kvm)
+	);
+
+
 #endif /* _TRACE_KVMS390_H */
 
 /* This part must be outside protection */
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 22859dc..72aa4a3 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -168,6 +168,7 @@ struct kvm_pit_config {
 #define KVM_EXIT_PAPR_HCALL	  19
 #define KVM_EXIT_S390_UCONTROL	  20
 #define KVM_EXIT_WATCHDOG         21
+#define KVM_EXIT_S390_TSCH        22
 
 /* For KVM_EXIT_INTERNAL_ERROR */
 /* Emulate instruction failed. */
@@ -285,6 +286,15 @@ struct kvm_run {
 			__u64 ret;
 			__u64 args[9];
 		} papr_hcall;
+		/* KVM_EXIT_S390_TSCH */
+		struct {
+			__u16 subchannel_id;
+			__u16 subchannel_nr;
+			__u32 io_int_parm;
+			__u32 io_int_word;
+			__u32 ipb;
+			__u8 dequeued;
+		} s390_tsch;
 		/* Fix the size of the union. */
 		char padding[256];
 	};
@@ -641,6 +651,7 @@ struct kvm_ppc_smmu_info {
 #endif
 #define KVM_CAP_IRQFD_RESAMPLE 82
 #define KVM_CAP_PPC_BOOKE_WATCHDOG 83
+#define KVM_CAP_S390_CSS_SUPPORT 84
 
 #ifdef KVM_CAP_IRQ_ROUTING
 
diff --git a/include/trace/events/kvm.h b/include/trace/events/kvm.h
index 7ef9e75..a23f47c 100644
--- a/include/trace/events/kvm.h
+++ b/include/trace/events/kvm.h
@@ -14,7 +14,7 @@
 	ERSN(SHUTDOWN), ERSN(FAIL_ENTRY), ERSN(INTR), ERSN(SET_TPR),	\
 	ERSN(TPR_ACCESS), ERSN(S390_SIEIC), ERSN(S390_RESET), ERSN(DCR),\
 	ERSN(NMI), ERSN(INTERNAL_ERROR), ERSN(OSI), ERSN(PAPR_HCALL),	\
-	ERSN(S390_UCONTROL)
+	ERSN(S390_UCONTROL), ERSN(S390_TSCH)
 
 TRACE_EVENT(kvm_userspace_exit,
 	    TP_PROTO(__u32 reason, int errno),
-- 
1.7.12.4

Re: [Qemu-devel] [RFC PATCH v3 0/5] s390: Host support for channel I/O.

[Marcelo Tosatti]

On Wed, Oct 31, 2012 at 05:24:33PM +0100, Cornelia Huck wrote:
> Hi,
> 
> here's the latest incarnation of my host patches to support channel
> I/O on s390.
> 
> Most patches have only seen minor fixes, but patch 5 is completely
> different since the kvm <-> user space interface has been reworked.
> 
> We now handle only interrupt-related operations in kvm. This
> includes two channel I/O instructions that can dequeue pending I/O
> interrupts: tpi and tsch (not the part actually interacting with
> the subchannel). This makes the interface less complex (only one
> new exit for tsch handling) and avoids duplicating code from qemu.
> 
> Cornelia Huck (5):
>   KVM: s390: Support for I/O interrupts.
>   KVM: s390: Add support for machine checks.
>   KVM: s390: In-kernel handling of I/O instructions.
>   KVM: s390: Base infrastructure for enabling capabilities.
>   KVM: s390: Add support for channel I/O instructions.
> 
>  Documentation/virtual/kvm/api.txt |  40 +++++-
>  arch/s390/include/asm/kvm_host.h  |  11 ++
>  arch/s390/kvm/intercept.c         |  22 ++-
>  arch/s390/kvm/interrupt.c         | 264 +++++++++++++++++++++++++++++++++++-
>  arch/s390/kvm/kvm-s390.c          |  38 ++++++
>  arch/s390/kvm/kvm-s390.h          |   6 +
>  arch/s390/kvm/priv.c              | 275 +++++++++++++++++++++++++++++++++++---
>  arch/s390/kvm/trace-s390.h        |  26 +++-
>  include/linux/kvm.h               |  18 +++
>  include/trace/events/kvm.h        |   2 +-
>  10 files changed, 673 insertions(+), 29 deletions(-)
> 
> -- 
> 1.7.12.4

Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com>