KVM: s390: Add support for machine checks.

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.
Reviewed-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Documentation/virtual/kvm/api.txt

@@ -2073,6 +2073,10 @@ 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.
 

arch/s390/include/asm/kvm_host.h

@@ -75,8 +75,10 @@ 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 */
@@ -187,6 +189,11 @@ 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;
@@ -197,6 +204,7 @@ 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;
 	};
 };
 

arch/s390/kvm/intercept.c

@@ -97,10 +97,12 @@ static int handle_lctl(struct kvm_vcpu *vcpu)
 
 static const 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,
 };

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;
 	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
 		if (psw_ioint_disabled(vcpu))
 			return 0;
@@ -116,6 +127,7 @@ 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)
@@ -139,6 +151,12 @@ 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;
 	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
 		if (psw_ioint_disabled(vcpu))
 			__set_cpuflag(vcpu, CPUSTAT_IO_INT);
@@ -326,6 +344,32 @@ 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;
+
 	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
 	{
 		__u32 param0 = ((__u32)inti->io.subchannel_id << 16) |
@@ -588,6 +632,61 @@ 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;
@@ -641,6 +740,13 @@ 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;
 	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
 		if (s390int->type & IOINT_AI_MASK)
 			VM_EVENT(kvm, 5, "%s", "inject: I/O (AI)");
@@ -749,6 +855,12 @@ 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:
 	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:

arch/s390/kvm/kvm-s390.h

@@ -106,6 +106,7 @@ 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,
@@ -117,6 +118,8 @@ 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);

arch/s390/kvm/priv.c

@@ -18,6 +18,8 @@
 #include <asm/debug.h>
 #include <asm/ebcdic.h>
 #include <asm/sysinfo.h>
+#include <asm/ptrace.h>
+#include <asm/compat.h>
 #include "gaccess.h"
 #include "kvm-s390.h"
 #include "trace.h"
@@ -166,6 +168,99 @@ 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);
+}
+
+#define PSW_MASK_ADDR_MODE (PSW_MASK_EA | PSW_MASK_BA)
+#define PSW_MASK_UNASSIGNED 0xb80800fe7fffffffUL
+#define PSW_ADDR_24 0x00000000000fffffUL
+#define PSW_ADDR_31 0x000000007fffffffUL
+
+int kvm_s390_handle_lpsw(struct kvm_vcpu *vcpu)
+{
+	u64 addr;
+	psw_compat_t new_psw;
+
+	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
+		return kvm_s390_inject_program_int(vcpu,
+						   PGM_PRIVILEGED_OPERATION);
+
+	addr = kvm_s390_get_base_disp_s(vcpu);
+
+	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;
+	}
+
+	if (!(new_psw.mask & PSW32_MASK_BASE)) {
+		kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+		goto out;
+	}
+
+	vcpu->arch.sie_block->gpsw.mask =
+		(new_psw.mask & ~PSW32_MASK_BASE) << 32;
+	vcpu->arch.sie_block->gpsw.addr = new_psw.addr;
+
+	if ((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_UNASSIGNED) ||
+	    (!(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_ADDR_MODE) &&
+	     (vcpu->arch.sie_block->gpsw.addr & ~PSW_ADDR_24)) ||
+	    ((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_ADDR_MODE) ==
+	     PSW_MASK_EA)) {
+		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)
+{
+	u64 addr;
+	psw_t new_psw;
+
+	addr = kvm_s390_get_base_disp_s(vcpu);
+
+	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 & PSW_MASK_UNASSIGNED) ||
+	    (((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_ADDR_MODE) ==
+	      PSW_MASK_BA) &&
+	     (vcpu->arch.sie_block->gpsw.addr & ~PSW_ADDR_31)) ||
+	    (!(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_ADDR_MODE) &&
+	     (vcpu->arch.sie_block->gpsw.addr & ~PSW_ADDR_24)) ||
+	    ((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_ADDR_MODE) ==
+	     PSW_MASK_EA)) {
+		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)
 {
 	u64 operand2;
@@ -292,6 +387,7 @@ static const 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)
@@ -316,6 +412,45 @@ 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;
+
+	/* This basically extracts the mask half of the psw. */
+	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 const 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)
 {
 	u64 address1, address2;

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)

include/uapi/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