Merge tag 'kvm-s390-next-4.8-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux into HEAD

KVM: s390: Features and fixes for 4.8 part1

Four bigger things:
1. The implementation of the STHYI opcode in the kernel. This is used
   in libraries like qclib [1] to provide enough information for a
   capacity and usage based software licence pricing. The STHYI content
   is defined by the related z/VM documentation [2]. Its data can be
   composed by accessing several other interfaces provided by LPAR or
   the machine. This information is partially sensitive or root-only
   so the kernel does the necessary filtering.
2. Preparation for nested virtualization (VSIE). KVM should query the
   proper sclp interfaces for the availability of some features before
   using it. In the past we have been sloppy and simply assumed that
   several features are available. With this we should be able to handle
   most cases of a missing feature.
3. CPU model interfaces extended by some additional features that are
   not covered by a facility bit in STFLE. For example all the crypto
   instructions of the coprocessor provide a query function. As reality
   tends to be more complex (e.g. export regulations might block some
   algorithms) we have to provide additional interfaces to query or
   set these non-stfle features.
4. Several fixes and changes detected and fixed when doing 1-3.

All features change base s390 code. All relevant patches have an ACK
from the s390 or component maintainers.

The next pull request for 4.8 (part2) will contain the implementation
of VSIE.

[1] http://www.ibm.com/developerworks/linux/linux390/qclib.html
[2] https://www.ibm.com/support/knowledgecenter/SSB27U_6.3.0/com.ibm.zvm.v630.hcpb4/hcpb4sth.htm

Documentation/virtual/kvm/api.txt

@@ -2520,6 +2520,7 @@ Parameters: struct kvm_device_attr
 Returns: 0 on success, -1 on error
 Errors:
   ENXIO:  The group or attribute is unknown/unsupported for this device
+          or hardware support is missing.
   EPERM:  The attribute cannot (currently) be accessed this way
           (e.g. read-only attribute, or attribute that only makes
           sense when the device is in a different state)
@@ -2547,6 +2548,7 @@ Parameters: struct kvm_device_attr
 Returns: 0 on success, -1 on error
 Errors:
   ENXIO:  The group or attribute is unknown/unsupported for this device
+          or hardware support is missing.
 
 Tests whether a device supports a particular attribute.  A successful
 return indicates the attribute is implemented.  It does not necessarily

Documentation/virtual/kvm/devices/vm.txt

@@ -20,7 +20,8 @@ Enables Collaborative Memory Management Assist (CMMA) for the virtual machine.
 
 1.2. ATTRIBUTE: KVM_S390_VM_MEM_CLR_CMMA
 Parameters: none
-Returns: 0
+Returns: -EINVAL if CMMA was not enabled
+         0 otherwise
 
 Clear the CMMA status for all guest pages, so any pages the guest marked
 as unused are again used any may not be reclaimed by the host.
@@ -85,6 +86,90 @@ Returns:    -EBUSY in case 1 or more vcpus are already activated (only in write
 	    -ENOMEM if not enough memory is available to process the ioctl
 	    0 in case of success
 
+2.3. ATTRIBUTE: KVM_S390_VM_CPU_MACHINE_FEAT (r/o)
+
+Allows user space to retrieve available cpu features. A feature is available if
+provided by the hardware and supported by kvm. In theory, cpu features could
+even be completely emulated by kvm.
+
+struct kvm_s390_vm_cpu_feat {
+        __u64 feat[16]; # Bitmap (1 = feature available), MSB 0 bit numbering
+};
+
+Parameters: address of a buffer to load the feature list from.
+Returns:    -EFAULT if the given address is not accessible from kernel space.
+	    0 in case of success.
+
+2.4. ATTRIBUTE: KVM_S390_VM_CPU_PROCESSOR_FEAT (r/w)
+
+Allows user space to retrieve or change enabled cpu features for all VCPUs of a
+VM. Features that are not available cannot be enabled.
+
+See 2.3. for a description of the parameter struct.
+
+Parameters: address of a buffer to store/load the feature list from.
+Returns:    -EFAULT if the given address is not accessible from kernel space.
+	    -EINVAL if a cpu feature that is not available is to be enabled.
+	    -EBUSY if at least one VCPU has already been defined.
+	    0 in case of success.
+
+2.5. ATTRIBUTE: KVM_S390_VM_CPU_MACHINE_SUBFUNC (r/o)
+
+Allows user space to retrieve available cpu subfunctions without any filtering
+done by a set IBC. These subfunctions are indicated to the guest VCPU via
+query or "test bit" subfunctions and used e.g. by cpacf functions, plo and ptff.
+
+A subfunction block is only valid if KVM_S390_VM_CPU_MACHINE contains the
+STFL(E) bit introducing the affected instruction. If the affected instruction
+indicates subfunctions via a "query subfunction", the response block is
+contained in the returned struct. If the affected instruction
+indicates subfunctions via a "test bit" mechanism, the subfunction codes are
+contained in the returned struct in MSB 0 bit numbering.
+
+struct kvm_s390_vm_cpu_subfunc {
+       u8 plo[32];           # always valid (ESA/390 feature)
+       u8 ptff[16];          # valid with TOD-clock steering
+       u8 kmac[16];          # valid with Message-Security-Assist
+       u8 kmc[16];           # valid with Message-Security-Assist
+       u8 km[16];            # valid with Message-Security-Assist
+       u8 kimd[16];          # valid with Message-Security-Assist
+       u8 klmd[16];          # valid with Message-Security-Assist
+       u8 pckmo[16];         # valid with Message-Security-Assist-Extension 3
+       u8 kmctr[16];         # valid with Message-Security-Assist-Extension 4
+       u8 kmf[16];           # valid with Message-Security-Assist-Extension 4
+       u8 kmo[16];           # valid with Message-Security-Assist-Extension 4
+       u8 pcc[16];           # valid with Message-Security-Assist-Extension 4
+       u8 ppno[16];          # valid with Message-Security-Assist-Extension 5
+       u8 reserved[1824];    # reserved for future instructions
+};
+
+Parameters: address of a buffer to load the subfunction blocks from.
+Returns:    -EFAULT if the given address is not accessible from kernel space.
+	    0 in case of success.
+
+2.6. ATTRIBUTE: KVM_S390_VM_CPU_PROCESSOR_SUBFUNC (r/w)
+
+Allows user space to retrieve or change cpu subfunctions to be indicated for
+all VCPUs of a VM. This attribute will only be available if kernel and
+hardware support are in place.
+
+The kernel uses the configured subfunction blocks for indication to
+the guest. A subfunction block will only be used if the associated STFL(E) bit
+has not been disabled by user space (so the instruction to be queried is
+actually available for the guest).
+
+As long as no data has been written, a read will fail. The IBC will be used
+to determine available subfunctions in this case, this will guarantee backward
+compatibility.
+
+See 2.5. for a description of the parameter struct.
+
+Parameters: address of a buffer to store/load the subfunction blocks from.
+Returns:    -EFAULT if the given address is not accessible from kernel space.
+	    -EINVAL when reading, if there was no write yet.
+	    -EBUSY if at least one VCPU has already been defined.
+	    0 in case of success.
+
 3. GROUP: KVM_S390_VM_TOD
 Architectures: s390
 

arch/s390/hypfs/hypfs_diag.c

@@ -19,29 +19,10 @@
 #include <asm/ebcdic.h>
 #include "hypfs.h"
 
-#define LPAR_NAME_LEN 8		/* lpar name len in diag 204 data */
-#define CPU_NAME_LEN 16		/* type name len of cpus in diag224 name table */
 #define TMP_SIZE 64		/* size of temporary buffers */
 
 #define DBFS_D204_HDR_VERSION	0
 
-/* diag 204 subcodes */
-enum diag204_sc {
-	SUBC_STIB4 = 4,
-	SUBC_RSI = 5,
-	SUBC_STIB6 = 6,
-	SUBC_STIB7 = 7
-};
-
-/* The two available diag 204 data formats */
-enum diag204_format {
-	INFO_SIMPLE = 0,
-	INFO_EXT = 0x00010000
-};
-
-/* bit is set in flags, when physical cpu info is included in diag 204 data */
-#define LPAR_PHYS_FLG  0x80
-
 static char *diag224_cpu_names;			/* diag 224 name table */
 static enum diag204_sc diag204_store_sc;	/* used subcode for store */
 static enum diag204_format diag204_info_type;	/* used diag 204 data format */
@@ -53,7 +34,7 @@ static int diag204_buf_pages;		/* number of pages for diag204 data */
 static struct dentry *dbfs_d204_file;
 
 /*
- * DIAG 204 data structures and member access functions.
+ * DIAG 204 member access functions.
  *
  * Since we have two different diag 204 data formats for old and new s390
  * machines, we do not access the structs directly, but use getter functions for
@@ -62,302 +43,173 @@ static struct dentry *dbfs_d204_file;
 
 /* Time information block */
 
-struct info_blk_hdr {
-	__u8  npar;
-	__u8  flags;
-	__u16 tslice;
-	__u16 phys_cpus;
-	__u16 this_part;
-	__u64 curtod;
-} __attribute__ ((packed));
-
-struct x_info_blk_hdr {
-	__u8  npar;
-	__u8  flags;
-	__u16 tslice;
-	__u16 phys_cpus;
-	__u16 this_part;
-	__u64 curtod1;
-	__u64 curtod2;
-	char reserved[40];
-} __attribute__ ((packed));
-
 static inline int info_blk_hdr__size(enum diag204_format type)
 {
-	if (type == INFO_SIMPLE)
-		return sizeof(struct info_blk_hdr);
-	else /* INFO_EXT */
-		return sizeof(struct x_info_blk_hdr);
+	if (type == DIAG204_INFO_SIMPLE)
+		return sizeof(struct diag204_info_blk_hdr);
+	else /* DIAG204_INFO_EXT */
+		return sizeof(struct diag204_x_info_blk_hdr);
 }
 
 static inline __u8 info_blk_hdr__npar(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct info_blk_hdr *)hdr)->npar;
-	else /* INFO_EXT */
-		return ((struct x_info_blk_hdr *)hdr)->npar;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_info_blk_hdr *)hdr)->npar;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_info_blk_hdr *)hdr)->npar;
 }
 
 static inline __u8 info_blk_hdr__flags(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct info_blk_hdr *)hdr)->flags;
-	else /* INFO_EXT */
-		return ((struct x_info_blk_hdr *)hdr)->flags;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_info_blk_hdr *)hdr)->flags;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_info_blk_hdr *)hdr)->flags;
 }
 
 static inline __u16 info_blk_hdr__pcpus(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct info_blk_hdr *)hdr)->phys_cpus;
-	else /* INFO_EXT */
-		return ((struct x_info_blk_hdr *)hdr)->phys_cpus;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_info_blk_hdr *)hdr)->phys_cpus;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_info_blk_hdr *)hdr)->phys_cpus;
 }
 
 /* Partition header */
 
-struct part_hdr {
-	__u8 pn;
-	__u8 cpus;
-	char reserved[6];
-	char part_name[LPAR_NAME_LEN];
-} __attribute__ ((packed));
-
-struct x_part_hdr {
-	__u8  pn;
-	__u8  cpus;
-	__u8  rcpus;
-	__u8  pflag;
-	__u32 mlu;
-	char  part_name[LPAR_NAME_LEN];
-	char  lpc_name[8];
-	char  os_name[8];
-	__u64 online_cs;
-	__u64 online_es;
-	__u8  upid;
-	char  reserved1[3];
-	__u32 group_mlu;
-	char  group_name[8];
-	char  reserved2[32];
-} __attribute__ ((packed));
-
 static inline int part_hdr__size(enum diag204_format type)
 {
-	if (type == INFO_SIMPLE)
-		return sizeof(struct part_hdr);
-	else /* INFO_EXT */
-		return sizeof(struct x_part_hdr);
+	if (type == DIAG204_INFO_SIMPLE)
+		return sizeof(struct diag204_part_hdr);
+	else /* DIAG204_INFO_EXT */
+		return sizeof(struct diag204_x_part_hdr);
 }
 
 static inline __u8 part_hdr__rcpus(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct part_hdr *)hdr)->cpus;
-	else /* INFO_EXT */
-		return ((struct x_part_hdr *)hdr)->rcpus;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_part_hdr *)hdr)->cpus;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_part_hdr *)hdr)->rcpus;
 }
 
 static inline void part_hdr__part_name(enum diag204_format type, void *hdr,
 				       char *name)
 {
-	if (type == INFO_SIMPLE)
-		memcpy(name, ((struct part_hdr *)hdr)->part_name,
-		       LPAR_NAME_LEN);
-	else /* INFO_EXT */
-		memcpy(name, ((struct x_part_hdr *)hdr)->part_name,
-		       LPAR_NAME_LEN);
-	EBCASC(name, LPAR_NAME_LEN);
-	name[LPAR_NAME_LEN] = 0;
+	if (type == DIAG204_INFO_SIMPLE)
+		memcpy(name, ((struct diag204_part_hdr *)hdr)->part_name,
+		       DIAG204_LPAR_NAME_LEN);
+	else /* DIAG204_INFO_EXT */
+		memcpy(name, ((struct diag204_x_part_hdr *)hdr)->part_name,
+		       DIAG204_LPAR_NAME_LEN);
+	EBCASC(name, DIAG204_LPAR_NAME_LEN);
+	name[DIAG204_LPAR_NAME_LEN] = 0;
 	strim(name);
 }
 
-struct cpu_info {
-	__u16 cpu_addr;
-	char  reserved1[2];
-	__u8  ctidx;
-	__u8  cflag;
-	__u16 weight;
-	__u64 acc_time;
-	__u64 lp_time;
-} __attribute__ ((packed));
-
-struct x_cpu_info {
-	__u16 cpu_addr;
-	char  reserved1[2];
-	__u8  ctidx;
-	__u8  cflag;
-	__u16 weight;
-	__u64 acc_time;
-	__u64 lp_time;
-	__u16 min_weight;
-	__u16 cur_weight;
-	__u16 max_weight;
-	char  reseved2[2];
-	__u64 online_time;
-	__u64 wait_time;
-	__u32 pma_weight;
-	__u32 polar_weight;
-	char  reserved3[40];
-} __attribute__ ((packed));
-
 /* CPU info block */
 
 static inline int cpu_info__size(enum diag204_format type)
 {
-	if (type == INFO_SIMPLE)
-		return sizeof(struct cpu_info);
-	else /* INFO_EXT */
-		return sizeof(struct x_cpu_info);
+	if (type == DIAG204_INFO_SIMPLE)
+		return sizeof(struct diag204_cpu_info);
+	else /* DIAG204_INFO_EXT */
+		return sizeof(struct diag204_x_cpu_info);
 }
 
 static inline __u8 cpu_info__ctidx(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct cpu_info *)hdr)->ctidx;
-	else /* INFO_EXT */
-		return ((struct x_cpu_info *)hdr)->ctidx;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_cpu_info *)hdr)->ctidx;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_cpu_info *)hdr)->ctidx;
 }
 
 static inline __u16 cpu_info__cpu_addr(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct cpu_info *)hdr)->cpu_addr;
-	else /* INFO_EXT */
-		return ((struct x_cpu_info *)hdr)->cpu_addr;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_cpu_info *)hdr)->cpu_addr;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_cpu_info *)hdr)->cpu_addr;
 }
 
 static inline __u64 cpu_info__acc_time(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct cpu_info *)hdr)->acc_time;
-	else /* INFO_EXT */
-		return ((struct x_cpu_info *)hdr)->acc_time;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_cpu_info *)hdr)->acc_time;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_cpu_info *)hdr)->acc_time;
 }
 
 static inline __u64 cpu_info__lp_time(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct cpu_info *)hdr)->lp_time;
-	else /* INFO_EXT */
-		return ((struct x_cpu_info *)hdr)->lp_time;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_cpu_info *)hdr)->lp_time;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_cpu_info *)hdr)->lp_time;
 }
 
 static inline __u64 cpu_info__online_time(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
+	if (type == DIAG204_INFO_SIMPLE)
 		return 0;	/* online_time not available in simple info */
-	else /* INFO_EXT */
-		return ((struct x_cpu_info *)hdr)->online_time;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_cpu_info *)hdr)->online_time;
 }
 
 /* Physical header */
 
-struct phys_hdr {
-	char reserved1[1];
-	__u8 cpus;
-	char reserved2[6];
-	char mgm_name[8];
-} __attribute__ ((packed));
-
-struct x_phys_hdr {
-	char reserved1[1];
-	__u8 cpus;
-	char reserved2[6];
-	char mgm_name[8];
-	char reserved3[80];
-} __attribute__ ((packed));
-
 static inline int phys_hdr__size(enum diag204_format type)
 {
-	if (type == INFO_SIMPLE)
-		return sizeof(struct phys_hdr);
-	else /* INFO_EXT */
-		return sizeof(struct x_phys_hdr);
+	if (type == DIAG204_INFO_SIMPLE)
+		return sizeof(struct diag204_phys_hdr);
+	else /* DIAG204_INFO_EXT */
+		return sizeof(struct diag204_x_phys_hdr);
 }
 
 static inline __u8 phys_hdr__cpus(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct phys_hdr *)hdr)->cpus;
-	else /* INFO_EXT */
-		return ((struct x_phys_hdr *)hdr)->cpus;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_phys_hdr *)hdr)->cpus;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_phys_hdr *)hdr)->cpus;
 }
 
 /* Physical CPU info block */
 
-struct phys_cpu {
-	__u16 cpu_addr;
-	char  reserved1[2];
-	__u8  ctidx;
-	char  reserved2[3];
-	__u64 mgm_time;
-	char  reserved3[8];
-} __attribute__ ((packed));
-
-struct x_phys_cpu {
-	__u16 cpu_addr;
-	char  reserved1[2];
-	__u8  ctidx;
-	char  reserved2[3];
-	__u64 mgm_time;
-	char  reserved3[80];
-} __attribute__ ((packed));
-
 static inline int phys_cpu__size(enum diag204_format type)
 {
-	if (type == INFO_SIMPLE)
-		return sizeof(struct phys_cpu);
-	else /* INFO_EXT */
-		return sizeof(struct x_phys_cpu);
+	if (type == DIAG204_INFO_SIMPLE)
+		return sizeof(struct diag204_phys_cpu);
+	else /* DIAG204_INFO_EXT */
+		return sizeof(struct diag204_x_phys_cpu);
 }
 
 static inline __u16 phys_cpu__cpu_addr(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct phys_cpu *)hdr)->cpu_addr;
-	else /* INFO_EXT */
-		return ((struct x_phys_cpu *)hdr)->cpu_addr;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_phys_cpu *)hdr)->cpu_addr;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_phys_cpu *)hdr)->cpu_addr;
 }
 
 static inline __u64 phys_cpu__mgm_time(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct phys_cpu *)hdr)->mgm_time;
-	else /* INFO_EXT */
-		return ((struct x_phys_cpu *)hdr)->mgm_time;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_phys_cpu *)hdr)->mgm_time;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_phys_cpu *)hdr)->mgm_time;
 }
 
 static inline __u64 phys_cpu__ctidx(enum diag204_format type, void *hdr)
 {
-	if (type == INFO_SIMPLE)
-		return ((struct phys_cpu *)hdr)->ctidx;
-	else /* INFO_EXT */
-		return ((struct x_phys_cpu *)hdr)->ctidx;
+	if (type == DIAG204_INFO_SIMPLE)
+		return ((struct diag204_phys_cpu *)hdr)->ctidx;
+	else /* DIAG204_INFO_EXT */
+		return ((struct diag204_x_phys_cpu *)hdr)->ctidx;
 }
 
 /* Diagnose 204 functions */
-
-static inline int __diag204(unsigned long subcode, unsigned long size, void *addr)
-{
-	register unsigned long _subcode asm("0") = subcode;
-	register unsigned long _size asm("1") = size;
-
-	asm volatile(
-		"	diag	%2,%0,0x204\n"
-		"0:\n"
-		EX_TABLE(0b,0b)
-		: "+d" (_subcode), "+d" (_size) : "d" (addr) : "memory");
-	if (_subcode)
-		return -1;
-	return _size;
-}
-
-static int diag204(unsigned long subcode, unsigned long size, void *addr)
-{
-	diag_stat_inc(DIAG_STAT_X204);
-	return __diag204(subcode, size, addr);
-}
-
 /*
  * For the old diag subcode 4 with simple data format we have to use real
  * memory. If we use subcode 6 or 7 with extended data format, we can (and
@@ -409,12 +261,12 @@ static void *diag204_get_buffer(enum diag204_format fmt, int *pages)
 		*pages = diag204_buf_pages;
 		return diag204_buf;
 	}
-	if (fmt == INFO_SIMPLE) {
+	if (fmt == DIAG204_INFO_SIMPLE) {
 		*pages = 1;
 		return diag204_alloc_rbuf();
-	} else {/* INFO_EXT */
-		*pages = diag204((unsigned long)SUBC_RSI |
-				 (unsigned long)INFO_EXT, 0, NULL);
+	} else {/* DIAG204_INFO_EXT */
+		*pages = diag204((unsigned long)DIAG204_SUBC_RSI |
+				 (unsigned long)DIAG204_INFO_EXT, 0, NULL);
 		if (*pages <= 0)
 			return ERR_PTR(-ENOSYS);
 		else
@@ -441,18 +293,18 @@ static int diag204_probe(void)
 	void *buf;
 	int pages, rc;
 
-	buf = diag204_get_buffer(INFO_EXT, &pages);
+	buf = diag204_get_buffer(DIAG204_INFO_EXT, &pages);
 	if (!IS_ERR(buf)) {
-		if (diag204((unsigned long)SUBC_STIB7 |
-			    (unsigned long)INFO_EXT, pages, buf) >= 0) {
-			diag204_store_sc = SUBC_STIB7;
-			diag204_info_type = INFO_EXT;
+		if (diag204((unsigned long)DIAG204_SUBC_STIB7 |
+			    (unsigned long)DIAG204_INFO_EXT, pages, buf) >= 0) {
+			diag204_store_sc = DIAG204_SUBC_STIB7;
+			diag204_info_type = DIAG204_INFO_EXT;
 			goto out;
 		}
-		if (diag204((unsigned long)SUBC_STIB6 |
-			    (unsigned long)INFO_EXT, pages, buf) >= 0) {
-			diag204_store_sc = SUBC_STIB6;
-			diag204_info_type = INFO_EXT;
+		if (diag204((unsigned long)DIAG204_SUBC_STIB6 |
+			    (unsigned long)DIAG204_INFO_EXT, pages, buf) >= 0) {
+			diag204_store_sc = DIAG204_SUBC_STIB6;
+			diag204_info_type = DIAG204_INFO_EXT;
 			goto out;
 		}
 		diag204_free_buffer();
@@ -460,15 +312,15 @@ static int diag204_probe(void)
 
 	/* subcodes 6 and 7 failed, now try subcode 4 */
 
-	buf = diag204_get_buffer(INFO_SIMPLE, &pages);
+	buf = diag204_get_buffer(DIAG204_INFO_SIMPLE, &pages);
 	if (IS_ERR(buf)) {
 		rc = PTR_ERR(buf);
 		goto fail_alloc;
 	}
-	if (diag204((unsigned long)SUBC_STIB4 |
-		    (unsigned long)INFO_SIMPLE, pages, buf) >= 0) {
-		diag204_store_sc = SUBC_STIB4;
-		diag204_info_type = INFO_SIMPLE;
+	if (diag204((unsigned long)DIAG204_SUBC_STIB4 |
+		    (unsigned long)DIAG204_INFO_SIMPLE, pages, buf) >= 0) {
+		diag204_store_sc = DIAG204_SUBC_STIB4;
+		diag204_info_type = DIAG204_INFO_SIMPLE;
 		goto out;
 	} else {
 		rc = -ENOSYS;
@@ -508,20 +360,6 @@ out:
 
 /* Diagnose 224 functions */
 
-static int diag224(void *ptr)
-{
-	int rc = -EOPNOTSUPP;
-
-	diag_stat_inc(DIAG_STAT_X224);
-	asm volatile(
-		"	diag	%1,%2,0x224\n"
-		"0:	lhi	%0,0x0\n"
-		"1:\n"
-		EX_TABLE(0b,1b)
-		: "+d" (rc) :"d" (0), "d" (ptr) : "memory");
-	return rc;
-}
-
 static int diag224_get_name_table(void)
 {
 	/* memory must be below 2GB */
@@ -543,9 +381,9 @@ static void diag224_delete_name_table(void)
 
 static int diag224_idx2name(int index, char *name)
 {
-	memcpy(name, diag224_cpu_names + ((index + 1) * CPU_NAME_LEN),
-		CPU_NAME_LEN);
-	name[CPU_NAME_LEN] = 0;
+	memcpy(name, diag224_cpu_names + ((index + 1) * DIAG204_CPU_NAME_LEN),
+	       DIAG204_CPU_NAME_LEN);
+	name[DIAG204_CPU_NAME_LEN] = 0;
 	strim(name);
 	return 0;
 }
@@ -601,7 +439,7 @@ __init int hypfs_diag_init(void)
 		pr_err("The hardware system does not support hypfs\n");
 		return -ENODATA;
 	}
-	if (diag204_info_type == INFO_EXT) {
+	if (diag204_info_type == DIAG204_INFO_EXT) {
 		rc = hypfs_dbfs_create_file(&dbfs_file_d204);
 		if (rc)
 			return rc;
@@ -649,7 +487,7 @@ static int hypfs_create_cpu_files(struct dentry *cpus_dir, void *cpu_info)
 			      cpu_info__lp_time(diag204_info_type, cpu_info));
 	if (IS_ERR(rc))
 		return PTR_ERR(rc);
-	if (diag204_info_type == INFO_EXT) {
+	if (diag204_info_type == DIAG204_INFO_EXT) {
 		rc = hypfs_create_u64(cpu_dir, "onlinetime",
 				      cpu_info__online_time(diag204_info_type,
 							    cpu_info));
@@ -665,12 +503,12 @@ static void *hypfs_create_lpar_files(struct dentry *systems_dir, void *part_hdr)
 {
 	struct dentry *cpus_dir;
 	struct dentry *lpar_dir;
-	char lpar_name[LPAR_NAME_LEN + 1];
+	char lpar_name[DIAG204_LPAR_NAME_LEN + 1];
 	void *cpu_info;
 	int i;
 
 	part_hdr__part_name(diag204_info_type, part_hdr, lpar_name);
-	lpar_name[LPAR_NAME_LEN] = 0;
+	lpar_name[DIAG204_LPAR_NAME_LEN] = 0;
 	lpar_dir = hypfs_mkdir(systems_dir, lpar_name);
 	if (IS_ERR(lpar_dir))
 		return lpar_dir;
@@ -753,7 +591,8 @@ int hypfs_diag_create_files(struct dentry *root)
 			goto err_out;
 		}
 	}
-	if (info_blk_hdr__flags(diag204_info_type, time_hdr) & LPAR_PHYS_FLG) {
+	if (info_blk_hdr__flags(diag204_info_type, time_hdr) &
+	    DIAG204_LPAR_PHYS_FLG) {
 		ptr = hypfs_create_phys_files(root, part_hdr);
 		if (IS_ERR(ptr)) {
 			rc = PTR_ERR(ptr);

arch/s390/include/asm/cpacf.h

@@ -20,6 +20,9 @@
 #define CPACF_KMC		0xb92f		/* MSA	*/
 #define CPACF_KIMD		0xb93e		/* MSA	*/
 #define CPACF_KLMD		0xb93f		/* MSA	*/
+#define CPACF_PCKMO		0xb928		/* MSA3 */
+#define CPACF_KMF		0xb92a		/* MSA4 */
+#define CPACF_KMO		0xb92b		/* MSA4 */
 #define CPACF_PCC		0xb92c		/* MSA4 */
 #define CPACF_KMCTR		0xb92d		/* MSA4 */
 #define CPACF_PPNO		0xb93c		/* MSA5 */
@@ -136,6 +139,7 @@ static inline void __cpacf_query(unsigned int opcode, unsigned char *status)
 	register unsigned long r1 asm("1") = (unsigned long) status;
 
 	asm volatile(
+		"	spm 0\n" /* pckmo doesn't change the cc */
 		/* Parameter registers are ignored, but may not be 0 */
 		"0:	.insn	rrf,%[opc] << 16,2,2,2,0\n"
 		"	brc	1,0b\n"	/* handle partial completion */
@@ -157,6 +161,12 @@ static inline int cpacf_query(unsigned int opcode, unsigned int func)
 		if (!test_facility(17))	/* check for MSA */
 			return 0;
 		break;
+	case CPACF_PCKMO:
+		if (!test_facility(76))	/* check for MSA3 */
+			return 0;
+		break;
+	case CPACF_KMF:
+	case CPACF_KMO:
 	case CPACF_PCC:
 	case CPACF_KMCTR:
 		if (!test_facility(77))	/* check for MSA4 */

arch/s390/include/asm/diag.h

@@ -78,4 +78,153 @@ struct diag210 {
 
 extern int diag210(struct diag210 *addr);
 
+/* bit is set in flags, when physical cpu info is included in diag 204 data */
+#define DIAG204_LPAR_PHYS_FLG 0x80
+#define DIAG204_LPAR_NAME_LEN 8		/* lpar name len in diag 204 data */
+#define DIAG204_CPU_NAME_LEN 16		/* type name len of cpus in diag224 name table */
+
+/* diag 204 subcodes */
+enum diag204_sc {
+	DIAG204_SUBC_STIB4 = 4,
+	DIAG204_SUBC_RSI = 5,
+	DIAG204_SUBC_STIB6 = 6,
+	DIAG204_SUBC_STIB7 = 7
+};
+
+/* The two available diag 204 data formats */
+enum diag204_format {
+	DIAG204_INFO_SIMPLE = 0,
+	DIAG204_INFO_EXT = 0x00010000
+};
+
+enum diag204_cpu_flags {
+	DIAG204_CPU_ONLINE = 0x20,
+	DIAG204_CPU_CAPPED = 0x40,
+};
+
+struct diag204_info_blk_hdr {
+	__u8  npar;
+	__u8  flags;
+	__u16 tslice;
+	__u16 phys_cpus;
+	__u16 this_part;
+	__u64 curtod;
+} __packed;
+
+struct diag204_x_info_blk_hdr {
+	__u8  npar;
+	__u8  flags;
+	__u16 tslice;
+	__u16 phys_cpus;
+	__u16 this_part;
+	__u64 curtod1;
+	__u64 curtod2;
+	char reserved[40];
+} __packed;
+
+struct diag204_part_hdr {
+	__u8 pn;
+	__u8 cpus;
+	char reserved[6];
+	char part_name[DIAG204_LPAR_NAME_LEN];
+} __packed;
+
+struct diag204_x_part_hdr {
+	__u8  pn;
+	__u8  cpus;
+	__u8  rcpus;
+	__u8  pflag;
+	__u32 mlu;
+	char  part_name[DIAG204_LPAR_NAME_LEN];
+	char  lpc_name[8];
+	char  os_name[8];
+	__u64 online_cs;
+	__u64 online_es;
+	__u8  upid;
+	__u8  reserved:3;
+	__u8  mtid:5;
+	char  reserved1[2];
+	__u32 group_mlu;
+	char  group_name[8];
+	char  hardware_group_name[8];
+	char  reserved2[24];
+} __packed;
+
+struct diag204_cpu_info {
+	__u16 cpu_addr;
+	char  reserved1[2];
+	__u8  ctidx;
+	__u8  cflag;
+	__u16 weight;
+	__u64 acc_time;
+	__u64 lp_time;
+} __packed;
+
+struct diag204_x_cpu_info {
+	__u16 cpu_addr;
+	char  reserved1[2];
+	__u8  ctidx;
+	__u8  cflag;
+	__u16 weight;
+	__u64 acc_time;
+	__u64 lp_time;
+	__u16 min_weight;
+	__u16 cur_weight;
+	__u16 max_weight;
+	char  reseved2[2];
+	__u64 online_time;
+	__u64 wait_time;
+	__u32 pma_weight;
+	__u32 polar_weight;
+	__u32 cpu_type_cap;
+	__u32 group_cpu_type_cap;
+	char  reserved3[32];
+} __packed;
+
+struct diag204_phys_hdr {
+	char reserved1[1];
+	__u8 cpus;
+	char reserved2[6];
+	char mgm_name[8];
+} __packed;
+
+struct diag204_x_phys_hdr {
+	char reserved1[1];
+	__u8 cpus;
+	char reserved2[6];
+	char mgm_name[8];
+	char reserved3[80];
+} __packed;
+
+struct diag204_phys_cpu {
+	__u16 cpu_addr;
+	char  reserved1[2];
+	__u8  ctidx;
+	char  reserved2[3];
+	__u64 mgm_time;
+	char  reserved3[8];
+} __packed;
+
+struct diag204_x_phys_cpu {
+	__u16 cpu_addr;
+	char  reserved1[2];
+	__u8  ctidx;
+	char  reserved2[1];
+	__u16 weight;
+	__u64 mgm_time;
+	char  reserved3[80];
+} __packed;
+
+struct diag204_x_part_block {
+	struct diag204_x_part_hdr hdr;
+	struct diag204_x_cpu_info cpus[];
+} __packed;
+
+struct diag204_x_phys_block {
+	struct diag204_x_phys_hdr hdr;
+	struct diag204_x_phys_cpu cpus[];
+} __packed;
+
+int diag204(unsigned long subcode, unsigned long size, void *addr);
+int diag224(void *ptr);
 #endif /* _ASM_S390_DIAG_H */

arch/s390/include/asm/kvm_host.h

@@ -154,6 +154,7 @@ struct kvm_s390_sie_block {
 #define LCTL_CR14	0x0002
 	__u16   lctl;			/* 0x0044 */
 	__s16	icpua;			/* 0x0046 */
+#define ICTL_OPEREXC	0x80000000
 #define ICTL_PINT	0x20000000
 #define ICTL_LPSW	0x00400000
 #define ICTL_STCTL	0x00040000
@@ -185,7 +186,9 @@ struct kvm_s390_sie_block {
 	__u32	scaol;			/* 0x0064 */
 	__u8	reserved68[4];		/* 0x0068 */
 	__u32	todpr;			/* 0x006c */
-	__u8	reserved70[32];		/* 0x0070 */
+	__u8	reserved70[16];		/* 0x0070 */
+	__u64	mso;			/* 0x0080 */
+	__u64	msl;			/* 0x0088 */
 	psw_t	gpsw;			/* 0x0090 */
 	__u64	gg14;			/* 0x00a0 */
 	__u64	gg15;			/* 0x00a8 */
@@ -255,6 +258,7 @@ struct kvm_vcpu_stat {
 	u32 instruction_stctg;
 	u32 exit_program_interruption;
 	u32 exit_instr_and_program;
+	u32 exit_operation_exception;
 	u32 deliver_external_call;
 	u32 deliver_emergency_signal;
 	u32 deliver_service_signal;
@@ -278,6 +282,7 @@ struct kvm_vcpu_stat {
 	u32 instruction_stfl;
 	u32 instruction_tprot;
 	u32 instruction_essa;
+	u32 instruction_sthyi;
 	u32 instruction_sigp_sense;
 	u32 instruction_sigp_sense_running;
 	u32 instruction_sigp_external_call;
@@ -649,11 +654,14 @@ struct kvm_arch{
 	wait_queue_head_t ipte_wq;
 	int ipte_lock_count;
 	struct mutex ipte_mutex;
+	struct ratelimit_state sthyi_limit;
 	spinlock_t start_stop_lock;
 	struct sie_page2 *sie_page2;
 	struct kvm_s390_cpu_model model;
 	struct kvm_s390_crypto crypto;
 	u64 epoch;
+	/* subset of available cpu features enabled by user space */
+	DECLARE_BITMAP(cpu_feat, KVM_S390_VM_CPU_FEAT_NR_BITS);
 };
 
 #define KVM_HVA_ERR_BAD		(-1UL)

arch/s390/include/asm/page.h

@@ -109,13 +109,14 @@ static inline unsigned char page_get_storage_key(unsigned long addr)
 
 static inline int page_reset_referenced(unsigned long addr)
 {
-	unsigned int ipm;
+	int cc;
 
 	asm volatile(
 		"	rrbe	0,%1\n"
 		"	ipm	%0\n"
-		: "=d" (ipm) : "a" (addr) : "cc");
-	return !!(ipm & 0x20000000);
+		"	srl	%0,28\n"
+		: "=d" (cc) : "a" (addr) : "cc");
+	return cc;
 }
 
 /* Bits int the storage key */

arch/s390/include/asm/pgtable.h

@@ -893,7 +893,12 @@ void ptep_zap_key(struct mm_struct *mm, unsigned long addr, pte_t *ptep);
 bool test_and_clear_guest_dirty(struct mm_struct *mm, unsigned long address);
 int set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 			  unsigned char key, bool nq);
-unsigned char get_guest_storage_key(struct mm_struct *mm, unsigned long addr);
+int cond_set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
+			       unsigned char key, unsigned char *oldkey,
+			       bool nq, bool mr, bool mc);
+int reset_guest_reference_bit(struct mm_struct *mm, unsigned long addr);
+int get_guest_storage_key(struct mm_struct *mm, unsigned long addr,
+			  unsigned char *key);
 
 /*
  * Certain architectures need to do special things when PTEs

arch/s390/include/asm/sclp.h

@@ -32,12 +32,19 @@ struct sclp_core_entry {
 	u8 reserved0;
 	u8 : 4;
 	u8 sief2 : 1;
-	u8 : 3;
-	u8 : 3;
+	u8 skey : 1;
+	u8 : 2;
+	u8 : 2;
+	u8 gpere : 1;
 	u8 siif : 1;
 	u8 sigpif : 1;
 	u8 : 3;
-	u8 reserved2[10];
+	u8 reserved2[3];
+	u8 : 2;
+	u8 ib : 1;
+	u8 cei : 1;
+	u8 : 4;
+	u8 reserved3[6];
 	u8 type;
 	u8 reserved1;
 } __attribute__((packed));
@@ -59,6 +66,15 @@ struct sclp_info {
 	unsigned char has_hvs : 1;
 	unsigned char has_esca : 1;
 	unsigned char has_sief2 : 1;
+	unsigned char has_64bscao : 1;
+	unsigned char has_gpere : 1;
+	unsigned char has_cmma : 1;
+	unsigned char has_gsls : 1;
+	unsigned char has_ib : 1;
+	unsigned char has_cei : 1;
+	unsigned char has_pfmfi : 1;
+	unsigned char has_ibs : 1;
+	unsigned char has_skey : 1;
 	unsigned int ibc;
 	unsigned int mtid;
 	unsigned int mtid_cp;
@@ -101,5 +117,6 @@ int memcpy_hsa_kernel(void *dest, unsigned long src, size_t count);
 int memcpy_hsa_user(void __user *dest, unsigned long src, size_t count);
 void sclp_early_detect(void);
 void _sclp_print_early(const char *);
+void sclp_ocf_cpc_name_copy(char *dst);
 
 #endif /* _ASM_S390_SCLP_H */

arch/s390/include/uapi/asm/kvm.h

@@ -93,6 +93,35 @@ struct kvm_s390_vm_cpu_machine {
 	__u64 fac_list[256];
 };
 
+#define KVM_S390_VM_CPU_PROCESSOR_FEAT	2
+#define KVM_S390_VM_CPU_MACHINE_FEAT	3
+
+#define KVM_S390_VM_CPU_FEAT_NR_BITS	1024
+#define KVM_S390_VM_CPU_FEAT_ESOP	0
+struct kvm_s390_vm_cpu_feat {
+	__u64 feat[16];
+};
+
+#define KVM_S390_VM_CPU_PROCESSOR_SUBFUNC	4
+#define KVM_S390_VM_CPU_MACHINE_SUBFUNC		5
+/* for "test bit" instructions MSB 0 bit ordering, for "query" raw blocks */
+struct kvm_s390_vm_cpu_subfunc {
+	__u8 plo[32];		/* always */
+	__u8 ptff[16];		/* with TOD-clock steering */
+	__u8 kmac[16];		/* with MSA */
+	__u8 kmc[16];		/* with MSA */
+	__u8 km[16];		/* with MSA */
+	__u8 kimd[16];		/* with MSA */
+	__u8 klmd[16];		/* with MSA */
+	__u8 pckmo[16];		/* with MSA3 */
+	__u8 kmctr[16];		/* with MSA4 */
+	__u8 kmf[16];		/* with MSA4 */
+	__u8 kmo[16];		/* with MSA4 */
+	__u8 pcc[16];		/* with MSA4 */
+	__u8 ppno[16];		/* with MSA5 */
+	__u8 reserved[1824];
+};
+
 /* kvm attributes for crypto */
 #define KVM_S390_VM_CRYPTO_ENABLE_AES_KW	0
 #define KVM_S390_VM_CRYPTO_ENABLE_DEA_KW	1

arch/s390/include/uapi/asm/sie.h

@@ -140,6 +140,7 @@
 	exit_code_ipa0(0xB2, 0x4c, "TAR"),	\
 	exit_code_ipa0(0xB2, 0x50, "CSP"),	\
 	exit_code_ipa0(0xB2, 0x54, "MVPG"),	\
+	exit_code_ipa0(0xB2, 0x56, "STHYI"),	\
 	exit_code_ipa0(0xB2, 0x58, "BSG"),	\
 	exit_code_ipa0(0xB2, 0x5a, "BSA"),	\
 	exit_code_ipa0(0xB2, 0x5f, "CHSC"),	\

arch/s390/kernel/diag.c

@@ -162,6 +162,28 @@ int diag14(unsigned long rx, unsigned long ry1, unsigned long subcode)
 }
 EXPORT_SYMBOL(diag14);
 
+static inline int __diag204(unsigned long subcode, unsigned long size, void *addr)
+{
+	register unsigned long _subcode asm("0") = subcode;
+	register unsigned long _size asm("1") = size;
+
+	asm volatile(
+		"	diag	%2,%0,0x204\n"
+		"0:\n"
+		EX_TABLE(0b,0b)
+		: "+d" (_subcode), "+d" (_size) : "d" (addr) : "memory");
+	if (_subcode)
+		return -1;
+	return _size;
+}
+
+int diag204(unsigned long subcode, unsigned long size, void *addr)
+{
+	diag_stat_inc(DIAG_STAT_X204);
+	return __diag204(subcode, size, addr);
+}
+EXPORT_SYMBOL(diag204);
+
 /*
  * Diagnose 210: Get information about a virtual device
  */
@@ -196,3 +218,18 @@ int diag210(struct diag210 *addr)
 	return ccode;
 }
 EXPORT_SYMBOL(diag210);
+
+int diag224(void *ptr)
+{
+	int rc = -EOPNOTSUPP;
+
+	diag_stat_inc(DIAG_STAT_X224);
+	asm volatile(
+		"	diag	%1,%2,0x224\n"
+		"0:	lhi	%0,0x0\n"
+		"1:\n"
+		EX_TABLE(0b,1b)
+		: "+d" (rc) :"d" (0), "d" (ptr) : "memory");
+	return rc;
+}
+EXPORT_SYMBOL(diag224);

arch/s390/kvm/Makefile

@@ -12,6 +12,6 @@ common-objs = $(KVM)/kvm_main.o $(KVM)/eventfd.o  $(KVM)/async_pf.o $(KVM)/irqch
 ccflags-y := -Ivirt/kvm -Iarch/s390/kvm
 
 kvm-objs := $(common-objs) kvm-s390.o intercept.o interrupt.o priv.o sigp.o
-kvm-objs += diag.o gaccess.o guestdbg.o
+kvm-objs += diag.o gaccess.o guestdbg.o sthyi.o
 
 obj-$(CONFIG_KVM) += kvm.o

arch/s390/kvm/diag.c

@@ -212,6 +212,11 @@ static int __diag_virtio_hypercall(struct kvm_vcpu *vcpu)
 	    (vcpu->run->s.regs.gprs[1] != KVM_S390_VIRTIO_CCW_NOTIFY))
 		return -EOPNOTSUPP;
 
+	VCPU_EVENT(vcpu, 4, "diag 0x500 schid 0x%8.8x queue 0x%x cookie 0x%llx",
+			    (u32) vcpu->run->s.regs.gprs[2],
+			    (u32) vcpu->run->s.regs.gprs[3],
+			    vcpu->run->s.regs.gprs[4]);
+
 	/*
 	 * The layout is as follows:
 	 * - gpr 2 contains the subchannel id (passed as addr)

arch/s390/kvm/gaccess.c

@@ -476,18 +476,73 @@ enum {
 	FSI_FETCH   = 2  /* Exception was due to fetch operation */
 };
 
-static int get_vcpu_asce(struct kvm_vcpu *vcpu, union asce *asce,
-			 ar_t ar, enum gacc_mode mode)
+enum prot_type {
+	PROT_TYPE_LA   = 0,
+	PROT_TYPE_KEYC = 1,
+	PROT_TYPE_ALC  = 2,
+	PROT_TYPE_DAT  = 3,
+};
+
+static int trans_exc(struct kvm_vcpu *vcpu, int code, unsigned long gva,
+		     ar_t ar, enum gacc_mode mode, enum prot_type prot)
 {
-	int rc;
-	struct psw_bits psw = psw_bits(vcpu->arch.sie_block->gpsw);
 	struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm;
-	struct trans_exc_code_bits *tec_bits;
+	struct trans_exc_code_bits *tec;
 
 	memset(pgm, 0, sizeof(*pgm));
-	tec_bits = (struct trans_exc_code_bits *)&pgm->trans_exc_code;
-	tec_bits->fsi = mode == GACC_STORE ? FSI_STORE : FSI_FETCH;
-	tec_bits->as = psw.as;
+	pgm->code = code;
+	tec = (struct trans_exc_code_bits *)&pgm->trans_exc_code;
+
+	switch (code) {
+	case PGM_ASCE_TYPE:
+	case PGM_PAGE_TRANSLATION:
+	case PGM_REGION_FIRST_TRANS:
+	case PGM_REGION_SECOND_TRANS:
+	case PGM_REGION_THIRD_TRANS:
+	case PGM_SEGMENT_TRANSLATION:
+		/*
+		 * op_access_id only applies to MOVE_PAGE -> set bit 61
+		 * exc_access_id has to be set to 0 for some instructions. Both
+		 * cases have to be handled by the caller. We can always store
+		 * exc_access_id, as it is undefined for non-ar cases.
+		 */
+		tec->addr = gva >> PAGE_SHIFT;
+		tec->fsi = mode == GACC_STORE ? FSI_STORE : FSI_FETCH;
+		tec->as = psw_bits(vcpu->arch.sie_block->gpsw).as;
+		/* FALL THROUGH */
+	case PGM_ALEN_TRANSLATION:
+	case PGM_ALE_SEQUENCE:
+	case PGM_ASTE_VALIDITY:
+	case PGM_ASTE_SEQUENCE:
+	case PGM_EXTENDED_AUTHORITY:
+		pgm->exc_access_id = ar;
+		break;
+	case PGM_PROTECTION:
+		switch (prot) {
+		case PROT_TYPE_ALC:
+			tec->b60 = 1;
+			/* FALL THROUGH */
+		case PROT_TYPE_DAT:
+			tec->b61 = 1;
+			tec->addr = gva >> PAGE_SHIFT;
+			tec->fsi = mode == GACC_STORE ? FSI_STORE : FSI_FETCH;
+			tec->as = psw_bits(vcpu->arch.sie_block->gpsw).as;
+			/* exc_access_id is undefined for most cases */
+			pgm->exc_access_id = ar;
+			break;
+		default: /* LA and KEYC set b61 to 0, other params undefined */
+			break;
+		}
+		break;
+	}
+	return code;
+}
+
+static int get_vcpu_asce(struct kvm_vcpu *vcpu, union asce *asce,
+			 unsigned long ga, ar_t ar, enum gacc_mode mode)
+{
+	int rc;
+	struct psw_bits psw = psw_bits(vcpu->arch.sie_block->gpsw);
 
 	if (!psw.t) {
 		asce->val = 0;
@@ -510,21 +565,8 @@ static int get_vcpu_asce(struct kvm_vcpu *vcpu, union asce *asce,
 		return 0;
 	case PSW_AS_ACCREG:
 		rc = ar_translation(vcpu, asce, ar, mode);
-		switch (rc) {
-		case PGM_ALEN_TRANSLATION:
-		case PGM_ALE_SEQUENCE:
-		case PGM_ASTE_VALIDITY:
-		case PGM_ASTE_SEQUENCE:
-		case PGM_EXTENDED_AUTHORITY:
-			vcpu->arch.pgm.exc_access_id = ar;
-			break;
-		case PGM_PROTECTION:
-			tec_bits->b60 = 1;
-			tec_bits->b61 = 1;
-			break;
-		}
 		if (rc > 0)
-			pgm->code = rc;
+			return trans_exc(vcpu, rc, ga, ar, mode, PROT_TYPE_ALC);
 		return rc;
 	}
 	return 0;
@@ -729,40 +771,31 @@ static int low_address_protection_enabled(struct kvm_vcpu *vcpu,
 	return 1;
 }
 
-static int guest_page_range(struct kvm_vcpu *vcpu, unsigned long ga,
+static int guest_page_range(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar,
 			    unsigned long *pages, unsigned long nr_pages,
 			    const union asce asce, enum gacc_mode mode)
 {
-	struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm;
 	psw_t *psw = &vcpu->arch.sie_block->gpsw;
-	struct trans_exc_code_bits *tec_bits;
-	int lap_enabled, rc;
+	int lap_enabled, rc = 0;
 
-	tec_bits = (struct trans_exc_code_bits *)&pgm->trans_exc_code;
 	lap_enabled = low_address_protection_enabled(vcpu, asce);
 	while (nr_pages) {
 		ga = kvm_s390_logical_to_effective(vcpu, ga);
-		tec_bits->addr = ga >> PAGE_SHIFT;
-		if (mode == GACC_STORE && lap_enabled && is_low_address(ga)) {
-			pgm->code = PGM_PROTECTION;
-			return pgm->code;
-		}
+		if (mode == GACC_STORE && lap_enabled && is_low_address(ga))
+			return trans_exc(vcpu, PGM_PROTECTION, ga, ar, mode,
+					 PROT_TYPE_LA);
 		ga &= PAGE_MASK;
 		if (psw_bits(*psw).t) {
 			rc = guest_translate(vcpu, ga, pages, asce, mode);
 			if (rc < 0)
 				return rc;
-			if (rc == PGM_PROTECTION)
-				tec_bits->b61 = 1;
-			if (rc)
-				pgm->code = rc;
 		} else {
 			*pages = kvm_s390_real_to_abs(vcpu, ga);
 			if (kvm_is_error_gpa(vcpu->kvm, *pages))
-				pgm->code = PGM_ADDRESSING;
+				rc = PGM_ADDRESSING;
 		}
-		if (pgm->code)
-			return pgm->code;
+		if (rc)
+			return trans_exc(vcpu, rc, ga, ar, mode, PROT_TYPE_DAT);
 		ga += PAGE_SIZE;
 		pages++;
 		nr_pages--;
@@ -783,7 +816,8 @@ int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data,
 
 	if (!len)
 		return 0;
-	rc = get_vcpu_asce(vcpu, &asce, ar, mode);
+	ga = kvm_s390_logical_to_effective(vcpu, ga);
+	rc = get_vcpu_asce(vcpu, &asce, ga, ar, mode);
 	if (rc)
 		return rc;
 	nr_pages = (((ga & ~PAGE_MASK) + len - 1) >> PAGE_SHIFT) + 1;
@@ -795,7 +829,7 @@ int access_guest(struct kvm_vcpu *vcpu, unsigned long ga, ar_t ar, void *data,
 	need_ipte_lock = psw_bits(*psw).t && !asce.r;
 	if (need_ipte_lock)
 		ipte_lock(vcpu);
-	rc = guest_page_range(vcpu, ga, pages, nr_pages, asce, mode);
+	rc = guest_page_range(vcpu, ga, ar, pages, nr_pages, asce, mode);
 	for (idx = 0; idx < nr_pages && !rc; idx++) {
 		gpa = *(pages + idx) + (ga & ~PAGE_MASK);
 		_len = min(PAGE_SIZE - (gpa & ~PAGE_MASK), len);
@@ -846,37 +880,28 @@ int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra,
 int guest_translate_address(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar,
 			    unsigned long *gpa, enum gacc_mode mode)
 {
-	struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm;
 	psw_t *psw = &vcpu->arch.sie_block->gpsw;
-	struct trans_exc_code_bits *tec;
 	union asce asce;
 	int rc;
 
 	gva = kvm_s390_logical_to_effective(vcpu, gva);
-	tec = (struct trans_exc_code_bits *)&pgm->trans_exc_code;
-	rc = get_vcpu_asce(vcpu, &asce, ar, mode);
-	tec->addr = gva >> PAGE_SHIFT;
+	rc = get_vcpu_asce(vcpu, &asce, gva, ar, mode);
 	if (rc)
 		return rc;
 	if (is_low_address(gva) && low_address_protection_enabled(vcpu, asce)) {
-		if (mode == GACC_STORE) {
-			rc = pgm->code = PGM_PROTECTION;
-			return rc;
-		}
+		if (mode == GACC_STORE)
+			return trans_exc(vcpu, PGM_PROTECTION, gva, 0,
+					 mode, PROT_TYPE_LA);
 	}
 
 	if (psw_bits(*psw).t && !asce.r) {	/* Use DAT? */
 		rc = guest_translate(vcpu, gva, gpa, asce, mode);
-		if (rc > 0) {
-			if (rc == PGM_PROTECTION)
-				tec->b61 = 1;
-			pgm->code = rc;
-		}
+		if (rc > 0)
+			return trans_exc(vcpu, rc, gva, 0, mode, PROT_TYPE_DAT);
 	} else {
-		rc = 0;
 		*gpa = kvm_s390_real_to_abs(vcpu, gva);
 		if (kvm_is_error_gpa(vcpu->kvm, *gpa))
-			rc = pgm->code = PGM_ADDRESSING;
+			return trans_exc(vcpu, rc, gva, PGM_ADDRESSING, mode, 0);
 	}
 
 	return rc;
@@ -915,20 +940,9 @@ int check_gva_range(struct kvm_vcpu *vcpu, unsigned long gva, ar_t ar,
  */
 int kvm_s390_check_low_addr_prot_real(struct kvm_vcpu *vcpu, unsigned long gra)
 {
-	struct kvm_s390_pgm_info *pgm = &vcpu->arch.pgm;
-	psw_t *psw = &vcpu->arch.sie_block->gpsw;
-	struct trans_exc_code_bits *tec_bits;
 	union ctlreg0 ctlreg0 = {.val = vcpu->arch.sie_block->gcr[0]};
 
 	if (!ctlreg0.lap || !is_low_address(gra))
 		return 0;
-
-	memset(pgm, 0, sizeof(*pgm));
-	tec_bits = (struct trans_exc_code_bits *)&pgm->trans_exc_code;
-	tec_bits->fsi = FSI_STORE;
-	tec_bits->as = psw_bits(*psw).as;
-	tec_bits->addr = gra >> PAGE_SHIFT;
-	pgm->code = PGM_PROTECTION;
-
-	return pgm->code;
+	return trans_exc(vcpu, PGM_PROTECTION, gra, 0, GACC_STORE, PROT_TYPE_LA);
 }

arch/s390/kvm/intercept.c

@@ -349,6 +349,19 @@ static int handle_partial_execution(struct kvm_vcpu *vcpu)
 	return -EOPNOTSUPP;
 }
 
+static int handle_operexc(struct kvm_vcpu *vcpu)
+{
+	vcpu->stat.exit_operation_exception++;
+	trace_kvm_s390_handle_operexc(vcpu, vcpu->arch.sie_block->ipa,
+				      vcpu->arch.sie_block->ipb);
+
+	if (vcpu->arch.sie_block->ipa == 0xb256 &&
+	    test_kvm_facility(vcpu->kvm, 74))
+		return handle_sthyi(vcpu);
+
+	return kvm_s390_inject_program_int(vcpu, PGM_OPERATION);
+}
+
 int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu)
 {
 	if (kvm_is_ucontrol(vcpu->kvm))
@@ -370,6 +383,8 @@ int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu)
 		return handle_validity(vcpu);
 	case 0x28:
 		return handle_stop(vcpu);
+	case 0x2c:
+		return handle_operexc(vcpu);
 	case 0x38:
 		return handle_partial_execution(vcpu);
 	default:

arch/s390/kvm/interrupt.c

@@ -28,9 +28,6 @@
 #include "gaccess.h"
 #include "trace-s390.h"
 
-#define IOINT_SCHID_MASK 0x0000ffff
-#define IOINT_SSID_MASK 0x00030000
-#define IOINT_CSSID_MASK 0x03fc0000
 #define PFAULT_INIT 0x0600
 #define PFAULT_DONE 0x0680
 #define VIRTIO_PARAM 0x0d00
@@ -821,7 +818,14 @@ static int __must_check __deliver_io(struct kvm_vcpu *vcpu,
 					struct kvm_s390_interrupt_info,
 					list);
 	if (inti) {
-		VCPU_EVENT(vcpu, 4, "deliver: I/O 0x%llx", inti->type);
+		if (inti->type & KVM_S390_INT_IO_AI_MASK)
+			VCPU_EVENT(vcpu, 4, "%s", "deliver: I/O (AI)");
+		else
+			VCPU_EVENT(vcpu, 4, "deliver: I/O %x ss %x schid %04x",
+			inti->io.subchannel_id >> 8,
+			inti->io.subchannel_id >> 1 & 0x3,
+			inti->io.subchannel_nr);
+
 		vcpu->stat.deliver_io_int++;
 		trace_kvm_s390_deliver_interrupt(vcpu->vcpu_id,
 				inti->type,
@@ -1415,6 +1419,13 @@ static int __inject_io(struct kvm *kvm, struct kvm_s390_interrupt_info *inti)
 	}
 	fi->counters[FIRQ_CNTR_IO] += 1;
 
+	if (inti->type & KVM_S390_INT_IO_AI_MASK)
+		VM_EVENT(kvm, 4, "%s", "inject: I/O (AI)");
+	else
+		VM_EVENT(kvm, 4, "inject: I/O %x ss %x schid %04x",
+			inti->io.subchannel_id >> 8,
+			inti->io.subchannel_id >> 1 & 0x3,
+			inti->io.subchannel_nr);
 	isc = int_word_to_isc(inti->io.io_int_word);
 	list = &fi->lists[FIRQ_LIST_IO_ISC_0 + isc];
 	list_add_tail(&inti->list, list);
@@ -1531,13 +1542,6 @@ int kvm_s390_inject_vm(struct kvm *kvm,
 		inti->mchk.mcic = s390int->parm64;
 		break;
 	case KVM_S390_INT_IO_MIN...KVM_S390_INT_IO_MAX:
-		if (inti->type & KVM_S390_INT_IO_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->io.subchannel_id = s390int->parm >> 16;
 		inti->io.subchannel_nr = s390int->parm & 0x0000ffffu;
 		inti->io.io_int_parm = s390int->parm64 >> 32;

arch/s390/kvm/kvm-s390.c

@@ -26,6 +26,7 @@
 #include <linux/slab.h>
 #include <linux/timer.h>
 #include <linux/vmalloc.h>
+#include <linux/bitmap.h>
 #include <asm/asm-offsets.h>
 #include <asm/lowcore.h>
 #include <asm/etr.h>
@@ -35,6 +36,8 @@
 #include <asm/switch_to.h>
 #include <asm/isc.h>
 #include <asm/sclp.h>
+#include <asm/cpacf.h>
+#include <asm/etr.h>
 #include "kvm-s390.h"
 #include "gaccess.h"
 
@@ -63,6 +66,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "exit_instruction", VCPU_STAT(exit_instruction) },
 	{ "exit_program_interruption", VCPU_STAT(exit_program_interruption) },
 	{ "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) },
+	{ "exit_operation_exception", VCPU_STAT(exit_operation_exception) },
 	{ "halt_successful_poll", VCPU_STAT(halt_successful_poll) },
 	{ "halt_attempted_poll", VCPU_STAT(halt_attempted_poll) },
 	{ "halt_poll_invalid", VCPU_STAT(halt_poll_invalid) },
@@ -93,6 +97,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
 	{ "instruction_stsi", VCPU_STAT(instruction_stsi) },
 	{ "instruction_stfl", VCPU_STAT(instruction_stfl) },
 	{ "instruction_tprot", VCPU_STAT(instruction_tprot) },
+	{ "instruction_sthyi", VCPU_STAT(instruction_sthyi) },
 	{ "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) },
 	{ "instruction_sigp_sense_running", VCPU_STAT(instruction_sigp_sense_running) },
 	{ "instruction_sigp_external_call", VCPU_STAT(instruction_sigp_external_call) },
@@ -130,6 +135,11 @@ unsigned long kvm_s390_fac_list_mask_size(void)
 	return ARRAY_SIZE(kvm_s390_fac_list_mask);
 }
 
+/* available cpu features supported by kvm */
+static DECLARE_BITMAP(kvm_s390_available_cpu_feat, KVM_S390_VM_CPU_FEAT_NR_BITS);
+/* available subfunctions indicated via query / "test bit" */
+static struct kvm_s390_vm_cpu_subfunc kvm_s390_available_subfunc;
+
 static struct gmap_notifier gmap_notifier;
 debug_info_t *kvm_s390_dbf;
 
@@ -187,6 +197,61 @@ void kvm_arch_hardware_unsetup(void)
 					 &kvm_clock_notifier);
 }
 
+static void allow_cpu_feat(unsigned long nr)
+{
+	set_bit_inv(nr, kvm_s390_available_cpu_feat);
+}
+
+static inline int plo_test_bit(unsigned char nr)
+{
+	register unsigned long r0 asm("0") = (unsigned long) nr | 0x100;
+	int cc = 3; /* subfunction not available */
+
+	asm volatile(
+		/* Parameter registers are ignored for "test bit" */
+		"	plo	0,0,0,0(0)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
+		: "=d" (cc)
+		: "d" (r0)
+		: "cc");
+	return cc == 0;
+}
+
+static void kvm_s390_cpu_feat_init(void)
+{
+	int i;
+
+	for (i = 0; i < 256; ++i) {
+		if (plo_test_bit(i))
+			kvm_s390_available_subfunc.plo[i >> 3] |= 0x80 >> (i & 7);
+	}
+
+	if (test_facility(28)) /* TOD-clock steering */
+		etr_ptff(kvm_s390_available_subfunc.ptff, ETR_PTFF_QAF);
+
+	if (test_facility(17)) { /* MSA */
+		__cpacf_query(CPACF_KMAC, kvm_s390_available_subfunc.kmac);
+		__cpacf_query(CPACF_KMC, kvm_s390_available_subfunc.kmc);
+		__cpacf_query(CPACF_KM, kvm_s390_available_subfunc.km);
+		__cpacf_query(CPACF_KIMD, kvm_s390_available_subfunc.kimd);
+		__cpacf_query(CPACF_KLMD, kvm_s390_available_subfunc.klmd);
+	}
+	if (test_facility(76)) /* MSA3 */
+		__cpacf_query(CPACF_PCKMO, kvm_s390_available_subfunc.pckmo);
+	if (test_facility(77)) { /* MSA4 */
+		__cpacf_query(CPACF_KMCTR, kvm_s390_available_subfunc.kmctr);
+		__cpacf_query(CPACF_KMF, kvm_s390_available_subfunc.kmf);
+		__cpacf_query(CPACF_KMO, kvm_s390_available_subfunc.kmo);
+		__cpacf_query(CPACF_PCC, kvm_s390_available_subfunc.pcc);
+	}
+	if (test_facility(57)) /* MSA5 */
+		__cpacf_query(CPACF_PPNO, kvm_s390_available_subfunc.ppno);
+
+	if (MACHINE_HAS_ESOP)
+		allow_cpu_feat(KVM_S390_VM_CPU_FEAT_ESOP);
+}
+
 int kvm_arch_init(void *opaque)
 {
 	kvm_s390_dbf = debug_register("kvm-trace", 32, 1, 7 * sizeof(long));
@@ -198,6 +263,8 @@ int kvm_arch_init(void *opaque)
 		return -ENOMEM;
 	}
 
+	kvm_s390_cpu_feat_init();
+
 	/* Register floating interrupt controller interface. */
 	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
 }
@@ -250,8 +317,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 		break;
 	case KVM_CAP_NR_VCPUS:
 	case KVM_CAP_MAX_VCPUS:
-		r = sclp.has_esca ? KVM_S390_ESCA_CPU_SLOTS
-				  : KVM_S390_BSCA_CPU_SLOTS;
+		r = KVM_S390_BSCA_CPU_SLOTS;
+		if (sclp.has_esca && sclp.has_64bscao)
+			r = KVM_S390_ESCA_CPU_SLOTS;
 		break;
 	case KVM_CAP_NR_MEMSLOTS:
 		r = KVM_USER_MEM_SLOTS;
@@ -417,9 +485,8 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att
 	unsigned int idx;
 	switch (attr->attr) {
 	case KVM_S390_VM_MEM_ENABLE_CMMA:
-		/* enable CMMA only for z10 and later (EDAT_1) */
-		ret = -EINVAL;
-		if (!MACHINE_IS_LPAR || !MACHINE_HAS_EDAT1)
+		ret = -ENXIO;
+		if (!sclp.has_cmma)
 			break;
 
 		ret = -EBUSY;
@@ -432,6 +499,9 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att
 		mutex_unlock(&kvm->lock);
 		break;
 	case KVM_S390_VM_MEM_CLR_CMMA:
+		ret = -ENXIO;
+		if (!sclp.has_cmma)
+			break;
 		ret = -EINVAL;
 		if (!kvm->arch.use_cmma)
 			break;
@@ -675,6 +745,39 @@ out:
 	return ret;
 }
 
+static int kvm_s390_set_processor_feat(struct kvm *kvm,
+				       struct kvm_device_attr *attr)
+{
+	struct kvm_s390_vm_cpu_feat data;
+	int ret = -EBUSY;
+
+	if (copy_from_user(&data, (void __user *)attr->addr, sizeof(data)))
+		return -EFAULT;
+	if (!bitmap_subset((unsigned long *) data.feat,
+			   kvm_s390_available_cpu_feat,
+			   KVM_S390_VM_CPU_FEAT_NR_BITS))
+		return -EINVAL;
+
+	mutex_lock(&kvm->lock);
+	if (!atomic_read(&kvm->online_vcpus)) {
+		bitmap_copy(kvm->arch.cpu_feat, (unsigned long *) data.feat,
+			    KVM_S390_VM_CPU_FEAT_NR_BITS);
+		ret = 0;
+	}
+	mutex_unlock(&kvm->lock);
+	return ret;
+}
+
+static int kvm_s390_set_processor_subfunc(struct kvm *kvm,
+					  struct kvm_device_attr *attr)
+{
+	/*
+	 * Once supported by kernel + hw, we have to store the subfunctions
+	 * in kvm->arch and remember that user space configured them.
+	 */
+	return -ENXIO;
+}
+
 static int kvm_s390_set_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr)
 {
 	int ret = -ENXIO;
@@ -683,6 +786,12 @@ static int kvm_s390_set_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr)
 	case KVM_S390_VM_CPU_PROCESSOR:
 		ret = kvm_s390_set_processor(kvm, attr);
 		break;
+	case KVM_S390_VM_CPU_PROCESSOR_FEAT:
+		ret = kvm_s390_set_processor_feat(kvm, attr);
+		break;
+	case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
+		ret = kvm_s390_set_processor_subfunc(kvm, attr);
+		break;
 	}
 	return ret;
 }
@@ -731,6 +840,50 @@ out:
 	return ret;
 }
 
+static int kvm_s390_get_processor_feat(struct kvm *kvm,
+				       struct kvm_device_attr *attr)
+{
+	struct kvm_s390_vm_cpu_feat data;
+
+	bitmap_copy((unsigned long *) data.feat, kvm->arch.cpu_feat,
+		    KVM_S390_VM_CPU_FEAT_NR_BITS);
+	if (copy_to_user((void __user *)attr->addr, &data, sizeof(data)))
+		return -EFAULT;
+	return 0;
+}
+
+static int kvm_s390_get_machine_feat(struct kvm *kvm,
+				     struct kvm_device_attr *attr)
+{
+	struct kvm_s390_vm_cpu_feat data;
+
+	bitmap_copy((unsigned long *) data.feat,
+		    kvm_s390_available_cpu_feat,
+		    KVM_S390_VM_CPU_FEAT_NR_BITS);
+	if (copy_to_user((void __user *)attr->addr, &data, sizeof(data)))
+		return -EFAULT;
+	return 0;
+}
+
+static int kvm_s390_get_processor_subfunc(struct kvm *kvm,
+					  struct kvm_device_attr *attr)
+{
+	/*
+	 * Once we can actually configure subfunctions (kernel + hw support),
+	 * we have to check if they were already set by user space, if so copy
+	 * them from kvm->arch.
+	 */
+	return -ENXIO;
+}
+
+static int kvm_s390_get_machine_subfunc(struct kvm *kvm,
+					struct kvm_device_attr *attr)
+{
+	if (copy_to_user((void __user *)attr->addr, &kvm_s390_available_subfunc,
+	    sizeof(struct kvm_s390_vm_cpu_subfunc)))
+		return -EFAULT;
+	return 0;
+}
 static int kvm_s390_get_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr)
 {
 	int ret = -ENXIO;
@@ -742,6 +895,18 @@ static int kvm_s390_get_cpu_model(struct kvm *kvm, struct kvm_device_attr *attr)
 	case KVM_S390_VM_CPU_MACHINE:
 		ret = kvm_s390_get_machine(kvm, attr);
 		break;
+	case KVM_S390_VM_CPU_PROCESSOR_FEAT:
+		ret = kvm_s390_get_processor_feat(kvm, attr);
+		break;
+	case KVM_S390_VM_CPU_MACHINE_FEAT:
+		ret = kvm_s390_get_machine_feat(kvm, attr);
+		break;
+	case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
+		ret = kvm_s390_get_processor_subfunc(kvm, attr);
+		break;
+	case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
+		ret = kvm_s390_get_machine_subfunc(kvm, attr);
+		break;
 	}
 	return ret;
 }
@@ -802,6 +967,8 @@ static int kvm_s390_vm_has_attr(struct kvm *kvm, struct kvm_device_attr *attr)
 		switch (attr->attr) {
 		case KVM_S390_VM_MEM_ENABLE_CMMA:
 		case KVM_S390_VM_MEM_CLR_CMMA:
+			ret = sclp.has_cmma ? 0 : -ENXIO;
+			break;
 		case KVM_S390_VM_MEM_LIMIT_SIZE:
 			ret = 0;
 			break;
@@ -825,8 +992,13 @@ static int kvm_s390_vm_has_attr(struct kvm *kvm, struct kvm_device_attr *attr)
 		switch (attr->attr) {
 		case KVM_S390_VM_CPU_PROCESSOR:
 		case KVM_S390_VM_CPU_MACHINE:
+		case KVM_S390_VM_CPU_PROCESSOR_FEAT:
+		case KVM_S390_VM_CPU_MACHINE_FEAT:
+		case KVM_S390_VM_CPU_MACHINE_SUBFUNC:
 			ret = 0;
 			break;
+		/* configuring subfunctions is not supported yet */
+		case KVM_S390_VM_CPU_PROCESSOR_SUBFUNC:
 		default:
 			ret = -ENXIO;
 			break;
@@ -857,7 +1029,6 @@ static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
 {
 	uint8_t *keys;
 	uint64_t hva;
-	unsigned long curkey;
 	int i, r = 0;
 
 	if (args->flags != 0)
@@ -878,26 +1049,27 @@ static long kvm_s390_get_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
 	if (!keys)
 		return -ENOMEM;
 
+	down_read(&current->mm->mmap_sem);
 	for (i = 0; i < args->count; i++) {
 		hva = gfn_to_hva(kvm, args->start_gfn + i);
 		if (kvm_is_error_hva(hva)) {
 			r = -EFAULT;
-			goto out;
+			break;
 		}
 
-		curkey = get_guest_storage_key(current->mm, hva);
-		if (IS_ERR_VALUE(curkey)) {
-			r = curkey;
-			goto out;
-		}
-		keys[i] = curkey;
+		r = get_guest_storage_key(current->mm, hva, &keys[i]);
+		if (r)
+			break;
+	}
+	up_read(&current->mm->mmap_sem);
+
+	if (!r) {
+		r = copy_to_user((uint8_t __user *)args->skeydata_addr, keys,
+				 sizeof(uint8_t) * args->count);
+		if (r)
+			r = -EFAULT;
 	}
 
-	r = copy_to_user((uint8_t __user *)args->skeydata_addr, keys,
-			 sizeof(uint8_t) * args->count);
-	if (r)
-		r = -EFAULT;
-out:
 	kvfree(keys);
 	return r;
 }
@@ -934,24 +1106,25 @@ static long kvm_s390_set_skeys(struct kvm *kvm, struct kvm_s390_skeys *args)
 	if (r)
 		goto out;
 
+	down_read(&current->mm->mmap_sem);
 	for (i = 0; i < args->count; i++) {
 		hva = gfn_to_hva(kvm, args->start_gfn + i);
 		if (kvm_is_error_hva(hva)) {
 			r = -EFAULT;
-			goto out;
+			break;
 		}
 
 		/* Lowest order bit is reserved */
 		if (keys[i] & 0x01) {
 			r = -EINVAL;
-			goto out;
+			break;
 		}
 
-		r = set_guest_storage_key(current->mm, hva,
-					  (unsigned long)keys[i], 0);
+		r = set_guest_storage_key(current->mm, hva, keys[i], 0);
 		if (r)
-			goto out;
+			break;
 	}
+	up_read(&current->mm->mmap_sem);
 out:
 	kvfree(keys);
 	return r;
@@ -1128,6 +1301,7 @@ static void sca_dispose(struct kvm *kvm)
 
 int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 {
+	gfp_t alloc_flags = GFP_KERNEL;
 	int i, rc;
 	char debug_name[16];
 	static unsigned long sca_offset;
@@ -1149,9 +1323,13 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 
 	rc = -ENOMEM;
 
+	ratelimit_state_init(&kvm->arch.sthyi_limit, 5 * HZ, 500);
+
 	kvm->arch.use_esca = 0; /* start with basic SCA */
+	if (!sclp.has_64bscao)
+		alloc_flags |= GFP_DMA;
 	rwlock_init(&kvm->arch.sca_lock);
-	kvm->arch.sca = (struct bsca_block *) get_zeroed_page(GFP_KERNEL);
+	kvm->arch.sca = (struct bsca_block *) get_zeroed_page(alloc_flags);
 	if (!kvm->arch.sca)
 		goto out_err;
 	spin_lock(&kvm_lock);
@@ -1188,6 +1366,9 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	memcpy(kvm->arch.model.fac_list, kvm->arch.model.fac_mask,
 	       S390_ARCH_FAC_LIST_SIZE_BYTE);
 
+	set_kvm_facility(kvm->arch.model.fac_mask, 74);
+	set_kvm_facility(kvm->arch.model.fac_list, 74);
+
 	kvm->arch.model.cpuid = kvm_s390_get_initial_cpuid();
 	kvm->arch.model.ibc = sclp.ibc & 0x0fff;
 
@@ -1395,7 +1576,7 @@ static int sca_can_add_vcpu(struct kvm *kvm, unsigned int id)
 
 	if (id < KVM_S390_BSCA_CPU_SLOTS)
 		return true;
-	if (!sclp.has_esca)
+	if (!sclp.has_esca || !sclp.has_64bscao)
 		return false;
 
 	mutex_lock(&kvm->lock);
@@ -1657,15 +1838,21 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 
 	kvm_s390_vcpu_setup_model(vcpu);
 
-	vcpu->arch.sie_block->ecb = 0x02;
+	/* pgste_set_pte has special handling for !MACHINE_HAS_ESOP */
+	if (MACHINE_HAS_ESOP)
+		vcpu->arch.sie_block->ecb |= 0x02;
 	if (test_kvm_facility(vcpu->kvm, 9))
 		vcpu->arch.sie_block->ecb |= 0x04;
-	if (test_kvm_facility(vcpu->kvm, 50) && test_kvm_facility(vcpu->kvm, 73))
+	if (test_kvm_facility(vcpu->kvm, 73))
 		vcpu->arch.sie_block->ecb |= 0x10;
 
-	if (test_kvm_facility(vcpu->kvm, 8))
+	if (test_kvm_facility(vcpu->kvm, 8) && sclp.has_pfmfi)
 		vcpu->arch.sie_block->ecb2 |= 0x08;
-	vcpu->arch.sie_block->eca   = 0xC1002000U;
+	vcpu->arch.sie_block->eca = 0x1002000U;
+	if (sclp.has_cei)
+		vcpu->arch.sie_block->eca |= 0x80000000U;
+	if (sclp.has_ib)
+		vcpu->arch.sie_block->eca |= 0x40000000U;
 	if (sclp.has_siif)
 		vcpu->arch.sie_block->eca |= 1;
 	if (sclp.has_sigpif)
@@ -1678,6 +1865,8 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 	}
 	vcpu->arch.sie_block->riccbd = (unsigned long) &vcpu->run->s.regs.riccb;
 	vcpu->arch.sie_block->ictl |= ICTL_ISKE | ICTL_SSKE | ICTL_RRBE;
+	if (test_kvm_facility(vcpu->kvm, 74))
+		vcpu->arch.sie_block->ictl |= ICTL_OPEREXC;
 
 	if (vcpu->kvm->arch.use_cmma) {
 		rc = kvm_s390_vcpu_setup_cmma(vcpu);
@@ -1715,6 +1904,10 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
 	vcpu->arch.sie_block = &sie_page->sie_block;
 	vcpu->arch.sie_block->itdba = (unsigned long) &sie_page->itdb;
 
+	/* the real guest size will always be smaller than msl */
+	vcpu->arch.sie_block->mso = 0;
+	vcpu->arch.sie_block->msl = sclp.hamax;
+
 	vcpu->arch.sie_block->icpua = id;
 	spin_lock_init(&vcpu->arch.local_int.lock);
 	vcpu->arch.local_int.float_int = &kvm->arch.float_int;
@@ -2001,6 +2194,8 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
 
 	if (dbg->control & ~VALID_GUESTDBG_FLAGS)
 		return -EINVAL;
+	if (!sclp.has_gpere)
+		return -EINVAL;
 
 	if (dbg->control & KVM_GUESTDBG_ENABLE) {
 		vcpu->guest_debug = dbg->control;
@@ -2597,6 +2792,8 @@ static void __disable_ibs_on_all_vcpus(struct kvm *kvm)
 
 static void __enable_ibs_on_vcpu(struct kvm_vcpu *vcpu)
 {
+	if (!sclp.has_ibs)
+		return;
 	kvm_check_request(KVM_REQ_DISABLE_IBS, vcpu);
 	kvm_s390_sync_request(KVM_REQ_ENABLE_IBS, vcpu);
 }

arch/s390/kvm/kvm-s390.h

@@ -175,6 +175,12 @@ static inline int set_kvm_facility(u64 *fac_list, unsigned long nr)
 	return 0;
 }
 
+static inline int test_kvm_cpu_feat(struct kvm *kvm, unsigned long nr)
+{
+	WARN_ON_ONCE(nr >= KVM_S390_VM_CPU_FEAT_NR_BITS);
+	return test_bit_inv(nr, kvm->arch.cpu_feat);
+}
+
 /* are cpu states controlled by user space */
 static inline int kvm_s390_user_cpu_state_ctrl(struct kvm *kvm)
 {
@@ -250,6 +256,9 @@ int kvm_s390_handle_eb(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu);
 int kvm_s390_handle_sigp_pei(struct kvm_vcpu *vcpu);
 
+/* implemented in sthyi.c */
+int handle_sthyi(struct kvm_vcpu *vcpu);
+
 /* implemented in kvm-s390.c */
 void kvm_s390_set_tod_clock(struct kvm *kvm, u64 tod);
 long kvm_arch_fault_in_page(struct kvm_vcpu *vcpu, gpa_t gpa, int writable);

arch/s390/kvm/priv.c

@@ -27,6 +27,7 @@
 #include <asm/io.h>
 #include <asm/ptrace.h>
 #include <asm/compat.h>
+#include <asm/sclp.h>
 #include "gaccess.h"
 #include "kvm-s390.h"
 #include "trace.h"
@@ -152,30 +153,166 @@ static int handle_store_cpu_address(struct kvm_vcpu *vcpu)
 static int __skey_check_enable(struct kvm_vcpu *vcpu)
 {
 	int rc = 0;
+
+	trace_kvm_s390_skey_related_inst(vcpu);
 	if (!(vcpu->arch.sie_block->ictl & (ICTL_ISKE | ICTL_SSKE | ICTL_RRBE)))
 		return rc;
 
 	rc = s390_enable_skey();
-	VCPU_EVENT(vcpu, 3, "%s", "enabling storage keys for guest");
-	trace_kvm_s390_skey_related_inst(vcpu);
-	vcpu->arch.sie_block->ictl &= ~(ICTL_ISKE | ICTL_SSKE | ICTL_RRBE);
+	VCPU_EVENT(vcpu, 3, "enabling storage keys for guest: %d", rc);
+	if (!rc)
+		vcpu->arch.sie_block->ictl &= ~(ICTL_ISKE | ICTL_SSKE | ICTL_RRBE);
 	return rc;
 }
 
-
-static int handle_skey(struct kvm_vcpu *vcpu)
+static int try_handle_skey(struct kvm_vcpu *vcpu)
 {
-	int rc = __skey_check_enable(vcpu);
+	int rc;
 
+	vcpu->stat.instruction_storage_key++;
+	rc = __skey_check_enable(vcpu);
 	if (rc)
 		return rc;
-	vcpu->stat.instruction_storage_key++;
-
+	if (sclp.has_skey) {
+		/* with storage-key facility, SIE interprets it for us */
+		kvm_s390_retry_instr(vcpu);
+		VCPU_EVENT(vcpu, 4, "%s", "retrying storage key operation");
+		return -EAGAIN;
+	}
 	if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
 		return kvm_s390_inject_program_int(vcpu, PGM_PRIVILEGED_OP);
+	return 0;
+}
 
-	kvm_s390_retry_instr(vcpu);
-	VCPU_EVENT(vcpu, 4, "%s", "retrying storage key operation");
+static int handle_iske(struct kvm_vcpu *vcpu)
+{
+	unsigned long addr;
+	unsigned char key;
+	int reg1, reg2;
+	int rc;
+
+	rc = try_handle_skey(vcpu);
+	if (rc)
+		return rc != -EAGAIN ? rc : 0;
+
+	kvm_s390_get_regs_rre(vcpu, &reg1, &reg2);
+
+	addr = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK;
+	addr = kvm_s390_logical_to_effective(vcpu, addr);
+	addr = kvm_s390_real_to_abs(vcpu, addr);
+	addr = gfn_to_hva(vcpu->kvm, gpa_to_gfn(addr));
+	if (kvm_is_error_hva(addr))
+		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+
+	down_read(&current->mm->mmap_sem);
+	rc = get_guest_storage_key(current->mm, addr, &key);
+	up_read(&current->mm->mmap_sem);
+	if (rc)
+		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+	vcpu->run->s.regs.gprs[reg1] &= ~0xff;
+	vcpu->run->s.regs.gprs[reg1] |= key;
+	return 0;
+}
+
+static int handle_rrbe(struct kvm_vcpu *vcpu)
+{
+	unsigned long addr;
+	int reg1, reg2;
+	int rc;
+
+	rc = try_handle_skey(vcpu);
+	if (rc)
+		return rc != -EAGAIN ? rc : 0;
+
+	kvm_s390_get_regs_rre(vcpu, &reg1, &reg2);
+
+	addr = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK;
+	addr = kvm_s390_logical_to_effective(vcpu, addr);
+	addr = kvm_s390_real_to_abs(vcpu, addr);
+	addr = gfn_to_hva(vcpu->kvm, gpa_to_gfn(addr));
+	if (kvm_is_error_hva(addr))
+		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+
+	down_read(&current->mm->mmap_sem);
+	rc = reset_guest_reference_bit(current->mm, addr);
+	up_read(&current->mm->mmap_sem);
+	if (rc < 0)
+		return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+
+	kvm_s390_set_psw_cc(vcpu, rc);
+	return 0;
+}
+
+#define SSKE_NQ 0x8
+#define SSKE_MR 0x4
+#define SSKE_MC 0x2
+#define SSKE_MB 0x1
+static int handle_sske(struct kvm_vcpu *vcpu)
+{
+	unsigned char m3 = vcpu->arch.sie_block->ipb >> 28;
+	unsigned long start, end;
+	unsigned char key, oldkey;
+	int reg1, reg2;
+	int rc;
+
+	rc = try_handle_skey(vcpu);
+	if (rc)
+		return rc != -EAGAIN ? rc : 0;
+
+	if (!test_kvm_facility(vcpu->kvm, 8))
+		m3 &= ~SSKE_MB;
+	if (!test_kvm_facility(vcpu->kvm, 10))
+		m3 &= ~(SSKE_MC | SSKE_MR);
+	if (!test_kvm_facility(vcpu->kvm, 14))
+		m3 &= ~SSKE_NQ;
+
+	kvm_s390_get_regs_rre(vcpu, &reg1, &reg2);
+
+	key = vcpu->run->s.regs.gprs[reg1] & 0xfe;
+	start = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK;
+	start = kvm_s390_logical_to_effective(vcpu, start);
+	if (m3 & SSKE_MB) {
+		/* start already designates an absolute address */
+		end = (start + (1UL << 20)) & ~((1UL << 20) - 1);
+	} else {
+		start = kvm_s390_real_to_abs(vcpu, start);
+		end = start + PAGE_SIZE;
+	}
+
+	while (start != end) {
+		unsigned long addr = gfn_to_hva(vcpu->kvm, gpa_to_gfn(start));
+
+		if (kvm_is_error_hva(addr))
+			return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+
+		down_read(&current->mm->mmap_sem);
+		rc = cond_set_guest_storage_key(current->mm, addr, key, &oldkey,
+						m3 & SSKE_NQ, m3 & SSKE_MR,
+						m3 & SSKE_MC);
+		up_read(&current->mm->mmap_sem);
+		if (rc < 0)
+			return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
+		start += PAGE_SIZE;
+	};
+
+	if (m3 & (SSKE_MC | SSKE_MR)) {
+		if (m3 & SSKE_MB) {
+			/* skey in reg1 is unpredictable */
+			kvm_s390_set_psw_cc(vcpu, 3);
+		} else {
+			kvm_s390_set_psw_cc(vcpu, rc);
+			vcpu->run->s.regs.gprs[reg1] &= ~0xff00UL;
+			vcpu->run->s.regs.gprs[reg1] |= (u64) oldkey << 8;
+		}
+	}
+	if (m3 & SSKE_MB) {
+		if (psw_bits(vcpu->arch.sie_block->gpsw).eaba == PSW_AMODE_64BIT)
+			vcpu->run->s.regs.gprs[reg2] &= ~PAGE_MASK;
+		else
+			vcpu->run->s.regs.gprs[reg2] &= ~0xfffff000UL;
+		end = kvm_s390_logical_to_effective(vcpu, end);
+		vcpu->run->s.regs.gprs[reg2] |= end;
+	}
 	return 0;
 }
 
@@ -583,9 +720,9 @@ static const intercept_handler_t b2_handlers[256] = {
 	[0x11] = handle_store_prefix,
 	[0x12] = handle_store_cpu_address,
 	[0x21] = handle_ipte_interlock,
-	[0x29] = handle_skey,
-	[0x2a] = handle_skey,
-	[0x2b] = handle_skey,
+	[0x29] = handle_iske,
+	[0x2a] = handle_rrbe,
+	[0x2b] = handle_sske,
 	[0x2c] = handle_test_block,
 	[0x30] = handle_io_inst,
 	[0x31] = handle_io_inst,
@@ -654,8 +791,10 @@ static int handle_epsw(struct kvm_vcpu *vcpu)
 
 static int handle_pfmf(struct kvm_vcpu *vcpu)
 {
+	bool mr = false, mc = false, nq;
 	int reg1, reg2;
 	unsigned long start, end;
+	unsigned char key;
 
 	vcpu->stat.instruction_pfmf++;
 
@@ -675,15 +814,27 @@ static int handle_pfmf(struct kvm_vcpu *vcpu)
 	    !test_kvm_facility(vcpu->kvm, 14))
 		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
 
-	/* No support for conditional-SSKE */
-	if (vcpu->run->s.regs.gprs[reg1] & (PFMF_MR | PFMF_MC))
-		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+	/* Only provide conditional-SSKE support if enabled for the guest */
+	if (vcpu->run->s.regs.gprs[reg1] & PFMF_SK &&
+	    test_kvm_facility(vcpu->kvm, 10)) {
+		mr = vcpu->run->s.regs.gprs[reg1] & PFMF_MR;
+		mc = vcpu->run->s.regs.gprs[reg1] & PFMF_MC;
+	}
 
+	nq = vcpu->run->s.regs.gprs[reg1] & PFMF_NQ;
+	key = vcpu->run->s.regs.gprs[reg1] & PFMF_KEY;
 	start = vcpu->run->s.regs.gprs[reg2] & PAGE_MASK;
 	start = kvm_s390_logical_to_effective(vcpu, start);
 
+	if (vcpu->run->s.regs.gprs[reg1] & PFMF_CF) {
+		if (kvm_s390_check_low_addr_prot_real(vcpu, start))
+			return kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm);
+	}
+
 	switch (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) {
 	case 0x00000000:
+		/* only 4k frames specify a real address */
+		start = kvm_s390_real_to_abs(vcpu, start);
 		end = (start + (1UL << 12)) & ~((1UL << 12) - 1);
 		break;
 	case 0x00001000:
@@ -701,20 +852,11 @@ static int handle_pfmf(struct kvm_vcpu *vcpu)
 		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
 	}
 
-	if (vcpu->run->s.regs.gprs[reg1] & PFMF_CF) {
-		if (kvm_s390_check_low_addr_prot_real(vcpu, start))
-			return kvm_s390_inject_prog_irq(vcpu, &vcpu->arch.pgm);
-	}
-
-	while (start < end) {
-		unsigned long useraddr, abs_addr;
+	while (start != end) {
+		unsigned long useraddr;
 
 		/* Translate guest address to host address */
-		if ((vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) == 0)
-			abs_addr = kvm_s390_real_to_abs(vcpu, start);
-		else
-			abs_addr = start;
-		useraddr = gfn_to_hva(vcpu->kvm, gpa_to_gfn(abs_addr));
+		useraddr = gfn_to_hva(vcpu->kvm, gpa_to_gfn(start));
 		if (kvm_is_error_hva(useraddr))
 			return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
 
@@ -728,16 +870,25 @@ static int handle_pfmf(struct kvm_vcpu *vcpu)
 
 			if (rc)
 				return rc;
-			if (set_guest_storage_key(current->mm, useraddr,
-					vcpu->run->s.regs.gprs[reg1] & PFMF_KEY,
-					vcpu->run->s.regs.gprs[reg1] & PFMF_NQ))
+			down_read(&current->mm->mmap_sem);
+			rc = cond_set_guest_storage_key(current->mm, useraddr,
+							key, NULL, nq, mr, mc);
+			up_read(&current->mm->mmap_sem);
+			if (rc < 0)
 				return kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING);
 		}
 
 		start += PAGE_SIZE;
 	}
-	if (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC)
-		vcpu->run->s.regs.gprs[reg2] = end;
+	if (vcpu->run->s.regs.gprs[reg1] & PFMF_FSC) {
+		if (psw_bits(vcpu->arch.sie_block->gpsw).eaba == PSW_AMODE_64BIT) {
+			vcpu->run->s.regs.gprs[reg2] = end;
+		} else {
+			vcpu->run->s.regs.gprs[reg2] &= ~0xffffffffUL;
+			end = kvm_s390_logical_to_effective(vcpu, end);
+			vcpu->run->s.regs.gprs[reg2] |= end;
+		}
+	}
 	return 0;
 }
 

arch/s390/kvm/sthyi.c

@@ -0,0 +1,471 @@
+/*
+ * store hypervisor information instruction emulation functions.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ *
+ * Copyright IBM Corp. 2016
+ * Author(s): Janosch Frank <frankja@linux.vnet.ibm.com>
+ */
+#include <linux/kvm_host.h>
+#include <linux/errno.h>
+#include <linux/pagemap.h>
+#include <linux/vmalloc.h>
+#include <linux/ratelimit.h>
+
+#include <asm/kvm_host.h>
+#include <asm/asm-offsets.h>
+#include <asm/sclp.h>
+#include <asm/diag.h>
+#include <asm/sysinfo.h>
+#include <asm/ebcdic.h>
+
+#include "kvm-s390.h"
+#include "gaccess.h"
+#include "trace.h"
+
+#define DED_WEIGHT 0xffff
+/*
+ * CP and IFL as EBCDIC strings, SP/0x40 determines the end of string
+ * as they are justified with spaces.
+ */
+#define CP  0xc3d7404040404040UL
+#define IFL 0xc9c6d34040404040UL
+
+enum hdr_flags {
+	HDR_NOT_LPAR   = 0x10,
+	HDR_STACK_INCM = 0x20,
+	HDR_STSI_UNAV  = 0x40,
+	HDR_PERF_UNAV  = 0x80,
+};
+
+enum mac_validity {
+	MAC_NAME_VLD = 0x20,
+	MAC_ID_VLD   = 0x40,
+	MAC_CNT_VLD  = 0x80,
+};
+
+enum par_flag {
+	PAR_MT_EN = 0x80,
+};
+
+enum par_validity {
+	PAR_GRP_VLD  = 0x08,
+	PAR_ID_VLD   = 0x10,
+	PAR_ABS_VLD  = 0x20,
+	PAR_WGHT_VLD = 0x40,
+	PAR_PCNT_VLD  = 0x80,
+};
+
+struct hdr_sctn {
+	u8 infhflg1;
+	u8 infhflg2; /* reserved */
+	u8 infhval1; /* reserved */
+	u8 infhval2; /* reserved */
+	u8 reserved[3];
+	u8 infhygct;
+	u16 infhtotl;
+	u16 infhdln;
+	u16 infmoff;
+	u16 infmlen;
+	u16 infpoff;
+	u16 infplen;
+	u16 infhoff1;
+	u16 infhlen1;
+	u16 infgoff1;
+	u16 infglen1;
+	u16 infhoff2;
+	u16 infhlen2;
+	u16 infgoff2;
+	u16 infglen2;
+	u16 infhoff3;
+	u16 infhlen3;
+	u16 infgoff3;
+	u16 infglen3;
+	u8 reserved2[4];
+} __packed;
+
+struct mac_sctn {
+	u8 infmflg1; /* reserved */
+	u8 infmflg2; /* reserved */
+	u8 infmval1;
+	u8 infmval2; /* reserved */
+	u16 infmscps;
+	u16 infmdcps;
+	u16 infmsifl;
+	u16 infmdifl;
+	char infmname[8];
+	char infmtype[4];
+	char infmmanu[16];
+	char infmseq[16];
+	char infmpman[4];
+	u8 reserved[4];
+} __packed;
+
+struct par_sctn {
+	u8 infpflg1;
+	u8 infpflg2; /* reserved */
+	u8 infpval1;
+	u8 infpval2; /* reserved */
+	u16 infppnum;
+	u16 infpscps;
+	u16 infpdcps;
+	u16 infpsifl;
+	u16 infpdifl;
+	u16 reserved;
+	char infppnam[8];
+	u32 infpwbcp;
+	u32 infpabcp;
+	u32 infpwbif;
+	u32 infpabif;
+	char infplgnm[8];
+	u32 infplgcp;
+	u32 infplgif;
+} __packed;
+
+struct sthyi_sctns {
+	struct hdr_sctn hdr;
+	struct mac_sctn mac;
+	struct par_sctn par;
+} __packed;
+
+struct cpu_inf {
+	u64 lpar_cap;
+	u64 lpar_grp_cap;
+	u64 lpar_weight;
+	u64 all_weight;
+	int cpu_num_ded;
+	int cpu_num_shd;
+};
+
+struct lpar_cpu_inf {
+	struct cpu_inf cp;
+	struct cpu_inf ifl;
+};
+
+static inline u64 cpu_id(u8 ctidx, void *diag224_buf)
+{
+	return *((u64 *)(diag224_buf + (ctidx + 1) * DIAG204_CPU_NAME_LEN));
+}
+
+/*
+ * Scales the cpu capping from the lpar range to the one expected in
+ * sthyi data.
+ *
+ * diag204 reports a cap in hundredths of processor units.
+ * z/VM's range for one core is 0 - 0x10000.
+ */
+static u32 scale_cap(u32 in)
+{
+	return (0x10000 * in) / 100;
+}
+
+static void fill_hdr(struct sthyi_sctns *sctns)
+{
+	sctns->hdr.infhdln = sizeof(sctns->hdr);
+	sctns->hdr.infmoff = sizeof(sctns->hdr);
+	sctns->hdr.infmlen = sizeof(sctns->mac);
+	sctns->hdr.infplen = sizeof(sctns->par);
+	sctns->hdr.infpoff = sctns->hdr.infhdln + sctns->hdr.infmlen;
+	sctns->hdr.infhtotl = sctns->hdr.infpoff + sctns->hdr.infplen;
+}
+
+static void fill_stsi_mac(struct sthyi_sctns *sctns,
+			  struct sysinfo_1_1_1 *sysinfo)
+{
+	if (stsi(sysinfo, 1, 1, 1))
+		return;
+
+	sclp_ocf_cpc_name_copy(sctns->mac.infmname);
+
+	memcpy(sctns->mac.infmtype, sysinfo->type, sizeof(sctns->mac.infmtype));
+	memcpy(sctns->mac.infmmanu, sysinfo->manufacturer, sizeof(sctns->mac.infmmanu));
+	memcpy(sctns->mac.infmpman, sysinfo->plant, sizeof(sctns->mac.infmpman));
+	memcpy(sctns->mac.infmseq, sysinfo->sequence, sizeof(sctns->mac.infmseq));
+
+	sctns->mac.infmval1 |= MAC_ID_VLD | MAC_NAME_VLD;
+}
+
+static void fill_stsi_par(struct sthyi_sctns *sctns,
+			  struct sysinfo_2_2_2 *sysinfo)
+{
+	if (stsi(sysinfo, 2, 2, 2))
+		return;
+
+	sctns->par.infppnum = sysinfo->lpar_number;
+	memcpy(sctns->par.infppnam, sysinfo->name, sizeof(sctns->par.infppnam));
+
+	sctns->par.infpval1 |= PAR_ID_VLD;
+}
+
+static void fill_stsi(struct sthyi_sctns *sctns)
+{
+	void *sysinfo;
+
+	/* Errors are handled through the validity bits in the response. */
+	sysinfo = (void *)__get_free_page(GFP_KERNEL);
+	if (!sysinfo)
+		return;
+
+	fill_stsi_mac(sctns, sysinfo);
+	fill_stsi_par(sctns, sysinfo);
+
+	free_pages((unsigned long)sysinfo, 0);
+}
+
+static void fill_diag_mac(struct sthyi_sctns *sctns,
+			  struct diag204_x_phys_block *block,
+			  void *diag224_buf)
+{
+	int i;
+
+	for (i = 0; i < block->hdr.cpus; i++) {
+		switch (cpu_id(block->cpus[i].ctidx, diag224_buf)) {
+		case CP:
+			if (block->cpus[i].weight == DED_WEIGHT)
+				sctns->mac.infmdcps++;
+			else
+				sctns->mac.infmscps++;
+			break;
+		case IFL:
+			if (block->cpus[i].weight == DED_WEIGHT)
+				sctns->mac.infmdifl++;
+			else
+				sctns->mac.infmsifl++;
+			break;
+		}
+	}
+	sctns->mac.infmval1 |= MAC_CNT_VLD;
+}
+
+/* Returns a pointer to the the next partition block. */
+static struct diag204_x_part_block *lpar_cpu_inf(struct lpar_cpu_inf *part_inf,
+						 bool this_lpar,
+						 void *diag224_buf,
+						 struct diag204_x_part_block *block)
+{
+	int i, capped = 0, weight_cp = 0, weight_ifl = 0;
+	struct cpu_inf *cpu_inf;
+
+	for (i = 0; i < block->hdr.rcpus; i++) {
+		if (!(block->cpus[i].cflag & DIAG204_CPU_ONLINE))
+			continue;
+
+		switch (cpu_id(block->cpus[i].ctidx, diag224_buf)) {
+		case CP:
+			cpu_inf = &part_inf->cp;
+			if (block->cpus[i].cur_weight < DED_WEIGHT)
+				weight_cp |= block->cpus[i].cur_weight;
+			break;
+		case IFL:
+			cpu_inf = &part_inf->ifl;
+			if (block->cpus[i].cur_weight < DED_WEIGHT)
+				weight_ifl |= block->cpus[i].cur_weight;
+			break;
+		default:
+			continue;
+		}
+
+		if (!this_lpar)
+			continue;
+
+		capped |= block->cpus[i].cflag & DIAG204_CPU_CAPPED;
+		cpu_inf->lpar_cap |= block->cpus[i].cpu_type_cap;
+		cpu_inf->lpar_grp_cap |= block->cpus[i].group_cpu_type_cap;
+
+		if (block->cpus[i].weight == DED_WEIGHT)
+			cpu_inf->cpu_num_ded += 1;
+		else
+			cpu_inf->cpu_num_shd += 1;
+	}
+
+	if (this_lpar && capped) {
+		part_inf->cp.lpar_weight = weight_cp;
+		part_inf->ifl.lpar_weight = weight_ifl;
+	}
+	part_inf->cp.all_weight += weight_cp;
+	part_inf->ifl.all_weight += weight_ifl;
+	return (struct diag204_x_part_block *)&block->cpus[i];
+}
+
+static void fill_diag(struct sthyi_sctns *sctns)
+{
+	int i, r, pages;
+	bool this_lpar;
+	void *diag204_buf;
+	void *diag224_buf = NULL;
+	struct diag204_x_info_blk_hdr *ti_hdr;
+	struct diag204_x_part_block *part_block;
+	struct diag204_x_phys_block *phys_block;
+	struct lpar_cpu_inf lpar_inf = {};
+
+	/* Errors are handled through the validity bits in the response. */
+	pages = diag204((unsigned long)DIAG204_SUBC_RSI |
+			(unsigned long)DIAG204_INFO_EXT, 0, NULL);
+	if (pages <= 0)
+		return;
+
+	diag204_buf = vmalloc(PAGE_SIZE * pages);
+	if (!diag204_buf)
+		return;
+
+	r = diag204((unsigned long)DIAG204_SUBC_STIB7 |
+		    (unsigned long)DIAG204_INFO_EXT, pages, diag204_buf);
+	if (r < 0)
+		goto out;
+
+	diag224_buf = kmalloc(PAGE_SIZE, GFP_KERNEL | GFP_DMA);
+	if (!diag224_buf || diag224(diag224_buf))
+		goto out;
+
+	ti_hdr = diag204_buf;
+	part_block = diag204_buf + sizeof(*ti_hdr);
+
+	for (i = 0; i < ti_hdr->npar; i++) {
+		/*
+		 * For the calling lpar we also need to get the cpu
+		 * caps and weights. The time information block header
+		 * specifies the offset to the partition block of the
+		 * caller lpar, so we know when we process its data.
+		 */
+		this_lpar = (void *)part_block - diag204_buf == ti_hdr->this_part;
+		part_block = lpar_cpu_inf(&lpar_inf, this_lpar, diag224_buf,
+					  part_block);
+	}
+
+	phys_block = (struct diag204_x_phys_block *)part_block;
+	part_block = diag204_buf + ti_hdr->this_part;
+	if (part_block->hdr.mtid)
+		sctns->par.infpflg1 = PAR_MT_EN;
+
+	sctns->par.infpval1 |= PAR_GRP_VLD;
+	sctns->par.infplgcp = scale_cap(lpar_inf.cp.lpar_grp_cap);
+	sctns->par.infplgif = scale_cap(lpar_inf.ifl.lpar_grp_cap);
+	memcpy(sctns->par.infplgnm, part_block->hdr.hardware_group_name,
+	       sizeof(sctns->par.infplgnm));
+
+	sctns->par.infpscps = lpar_inf.cp.cpu_num_shd;
+	sctns->par.infpdcps = lpar_inf.cp.cpu_num_ded;
+	sctns->par.infpsifl = lpar_inf.ifl.cpu_num_shd;
+	sctns->par.infpdifl = lpar_inf.ifl.cpu_num_ded;
+	sctns->par.infpval1 |= PAR_PCNT_VLD;
+
+	sctns->par.infpabcp = scale_cap(lpar_inf.cp.lpar_cap);
+	sctns->par.infpabif = scale_cap(lpar_inf.ifl.lpar_cap);
+	sctns->par.infpval1 |= PAR_ABS_VLD;
+
+	/*
+	 * Everything below needs global performance data to be
+	 * meaningful.
+	 */
+	if (!(ti_hdr->flags & DIAG204_LPAR_PHYS_FLG)) {
+		sctns->hdr.infhflg1 |= HDR_PERF_UNAV;
+		goto out;
+	}
+
+	fill_diag_mac(sctns, phys_block, diag224_buf);
+
+	if (lpar_inf.cp.lpar_weight) {
+		sctns->par.infpwbcp = sctns->mac.infmscps * 0x10000 *
+			lpar_inf.cp.lpar_weight / lpar_inf.cp.all_weight;
+	}
+
+	if (lpar_inf.ifl.lpar_weight) {
+		sctns->par.infpwbif = sctns->mac.infmsifl * 0x10000 *
+			lpar_inf.ifl.lpar_weight / lpar_inf.ifl.all_weight;
+	}
+	sctns->par.infpval1 |= PAR_WGHT_VLD;
+
+out:
+	kfree(diag224_buf);
+	vfree(diag204_buf);
+}
+
+static int sthyi(u64 vaddr)
+{
+	register u64 code asm("0") = 0;
+	register u64 addr asm("2") = vaddr;
+	int cc;
+
+	asm volatile(
+		".insn   rre,0xB2560000,%[code],%[addr]\n"
+		"ipm     %[cc]\n"
+		"srl     %[cc],28\n"
+		: [cc] "=d" (cc)
+		: [code] "d" (code), [addr] "a" (addr)
+		: "memory", "cc");
+	return cc;
+}
+
+int handle_sthyi(struct kvm_vcpu *vcpu)
+{
+	int reg1, reg2, r = 0;
+	u64 code, addr, cc = 0;
+	struct sthyi_sctns *sctns = NULL;
+
+	/*
+	 * STHYI requires extensive locking in the higher hypervisors
+	 * and is very computational/memory expensive. Therefore we
+	 * ratelimit the executions per VM.
+	 */
+	if (!__ratelimit(&vcpu->kvm->arch.sthyi_limit)) {
+		kvm_s390_retry_instr(vcpu);
+		return 0;
+	}
+
+	kvm_s390_get_regs_rre(vcpu, &reg1, &reg2);
+	code = vcpu->run->s.regs.gprs[reg1];
+	addr = vcpu->run->s.regs.gprs[reg2];
+
+	vcpu->stat.instruction_sthyi++;
+	VCPU_EVENT(vcpu, 3, "STHYI: fc: %llu addr: 0x%016llx", code, addr);
+	trace_kvm_s390_handle_sthyi(vcpu, code, addr);
+
+	if (reg1 == reg2 || reg1 & 1 || reg2 & 1 || addr & ~PAGE_MASK)
+		return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION);
+
+	if (code & 0xffff) {
+		cc = 3;
+		goto out;
+	}
+
+	/*
+	 * If the page has not yet been faulted in, we want to do that
+	 * now and not after all the expensive calculations.
+	 */
+	r = write_guest(vcpu, addr, reg2, &cc, 1);
+	if (r)
+		return kvm_s390_inject_prog_cond(vcpu, r);
+
+	sctns = (void *)get_zeroed_page(GFP_KERNEL);
+	if (!sctns)
+		return -ENOMEM;
+
+	/*
+	 * If we are a guest, we don't want to emulate an emulated
+	 * instruction. We ask the hypervisor to provide the data.
+	 */
+	if (test_facility(74)) {
+		cc = sthyi((u64)sctns);
+		goto out;
+	}
+
+	fill_hdr(sctns);
+	fill_stsi(sctns);
+	fill_diag(sctns);
+
+out:
+	if (!cc) {
+		r = write_guest(vcpu, addr, reg2, sctns, PAGE_SIZE);
+		if (r) {
+			free_page((unsigned long)sctns);
+			return kvm_s390_inject_prog_cond(vcpu, r);
+		}
+	}
+
+	free_page((unsigned long)sctns);
+	vcpu->run->s.regs.gprs[reg2 + 1] = cc ? 4 : 0;
+	kvm_s390_set_psw_cc(vcpu, cc);
+	return r;
+}

arch/s390/kvm/trace.h

@@ -41,7 +41,7 @@ TRACE_EVENT(kvm_s390_skey_related_inst,
 	    TP_fast_assign(
 		    VCPU_ASSIGN_COMMON
 		    ),
-	    VCPU_TP_PRINTK("%s", "first instruction related to skeys on vcpu")
+	    VCPU_TP_PRINTK("%s", "storage key related instruction")
 	);
 
 TRACE_EVENT(kvm_s390_major_guest_pfault,
@@ -185,8 +185,10 @@ TRACE_EVENT(kvm_s390_intercept_prog,
 		    __entry->code = code;
 		    ),
 
-	    VCPU_TP_PRINTK("intercepted program interruption %04x",
-			   __entry->code)
+	    VCPU_TP_PRINTK("intercepted program interruption %04x (%s)",
+			   __entry->code,
+			   __print_symbolic(__entry->code,
+					    icpt_prog_codes))
 	);
 
 /*
@@ -412,6 +414,47 @@ TRACE_EVENT(kvm_s390_handle_stsi,
 			   __entry->addr)
 	);
 
+TRACE_EVENT(kvm_s390_handle_operexc,
+	    TP_PROTO(VCPU_PROTO_COMMON, __u16 ipa, __u32 ipb),
+	    TP_ARGS(VCPU_ARGS_COMMON, ipa, ipb),
+
+	    TP_STRUCT__entry(
+		    VCPU_FIELD_COMMON
+		    __field(__u64, instruction)
+		    ),
+
+	    TP_fast_assign(
+		    VCPU_ASSIGN_COMMON
+		    __entry->instruction = ((__u64)ipa << 48) |
+		    ((__u64)ipb << 16);
+		    ),
+
+	    VCPU_TP_PRINTK("operation exception on instruction %016llx (%s)",
+			   __entry->instruction,
+			   __print_symbolic(icpt_insn_decoder(__entry->instruction),
+					    icpt_insn_codes))
+	);
+
+TRACE_EVENT(kvm_s390_handle_sthyi,
+	    TP_PROTO(VCPU_PROTO_COMMON, u64 code, u64 addr),
+	    TP_ARGS(VCPU_ARGS_COMMON, code, addr),
+
+	    TP_STRUCT__entry(
+		    VCPU_FIELD_COMMON
+		    __field(u64, code)
+		    __field(u64, addr)
+		    ),
+
+	    TP_fast_assign(
+		    VCPU_ASSIGN_COMMON
+		    __entry->code = code;
+		    __entry->addr = addr;
+		    ),
+
+	    VCPU_TP_PRINTK("STHYI fc: %llu addr: %016llx",
+			   __entry->code, __entry->addr)
+	);
+
 #endif /* _TRACE_KVM_H */
 
 /* This part must be outside protection */

arch/s390/mm/pgtable.c

@@ -506,12 +506,9 @@ int set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 	pgste_t old, new;
 	pte_t *ptep;
 
-	down_read(&mm->mmap_sem);
 	ptep = get_locked_pte(mm, addr, &ptl);
-	if (unlikely(!ptep)) {
-		up_read(&mm->mmap_sem);
+	if (unlikely(!ptep))
 		return -EFAULT;
-	}
 
 	new = old = pgste_get_lock(ptep);
 	pgste_val(new) &= ~(PGSTE_GR_BIT | PGSTE_GC_BIT |
@@ -538,45 +535,100 @@ int set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
 
 	pgste_set_unlock(ptep, new);
 	pte_unmap_unlock(ptep, ptl);
-	up_read(&mm->mmap_sem);
 	return 0;
 }
 EXPORT_SYMBOL(set_guest_storage_key);
 
-unsigned char get_guest_storage_key(struct mm_struct *mm, unsigned long addr)
+/**
+ * Conditionally set a guest storage key (handling csske).
+ * oldkey will be updated when either mr or mc is set and a pointer is given.
+ *
+ * Returns 0 if a guests storage key update wasn't necessary, 1 if the guest
+ * storage key was updated and -EFAULT on access errors.
+ */
+int cond_set_guest_storage_key(struct mm_struct *mm, unsigned long addr,
+			       unsigned char key, unsigned char *oldkey,
+			       bool nq, bool mr, bool mc)
+{
+	unsigned char tmp, mask = _PAGE_ACC_BITS | _PAGE_FP_BIT;
+	int rc;
+
+	/* we can drop the pgste lock between getting and setting the key */
+	if (mr | mc) {
+		rc = get_guest_storage_key(current->mm, addr, &tmp);
+		if (rc)
+			return rc;
+		if (oldkey)
+			*oldkey = tmp;
+		if (!mr)
+			mask |= _PAGE_REFERENCED;
+		if (!mc)
+			mask |= _PAGE_CHANGED;
+		if (!((tmp ^ key) & mask))
+			return 0;
+	}
+	rc = set_guest_storage_key(current->mm, addr, key, nq);
+	return rc < 0 ? rc : 1;
+}
+EXPORT_SYMBOL(cond_set_guest_storage_key);
+
+/**
+ * Reset a guest reference bit (rrbe), returning the reference and changed bit.
+ *
+ * Returns < 0 in case of error, otherwise the cc to be reported to the guest.
+ */
+int reset_guest_reference_bit(struct mm_struct *mm, unsigned long addr)
 {
-	unsigned char key;
 	spinlock_t *ptl;
-	pgste_t pgste;
+	pgste_t old, new;
 	pte_t *ptep;
+	int cc = 0;
 
-	down_read(&mm->mmap_sem);
 	ptep = get_locked_pte(mm, addr, &ptl);
-	if (unlikely(!ptep)) {
-		up_read(&mm->mmap_sem);
+	if (unlikely(!ptep))
 		return -EFAULT;
-	}
-	pgste = pgste_get_lock(ptep);
 
-	if (pte_val(*ptep) & _PAGE_INVALID) {
-		key  = (pgste_val(pgste) & PGSTE_ACC_BITS) >> 56;
-		key |= (pgste_val(pgste) & PGSTE_FP_BIT) >> 56;
-		key |= (pgste_val(pgste) & PGSTE_GR_BIT) >> 48;
-		key |= (pgste_val(pgste) & PGSTE_GC_BIT) >> 48;
-	} else {
-		key = page_get_storage_key(pte_val(*ptep) & PAGE_MASK);
+	new = old = pgste_get_lock(ptep);
+	/* Reset guest reference bit only */
+	pgste_val(new) &= ~PGSTE_GR_BIT;
 
-		/* Reflect guest's logical view, not physical */
-		if (pgste_val(pgste) & PGSTE_GR_BIT)
-			key |= _PAGE_REFERENCED;
-		if (pgste_val(pgste) & PGSTE_GC_BIT)
-			key |= _PAGE_CHANGED;
+	if (!(pte_val(*ptep) & _PAGE_INVALID)) {
+		cc = page_reset_referenced(pte_val(*ptep) & PAGE_MASK);
+		/* Merge real referenced bit into host-set */
+		pgste_val(new) |= ((unsigned long) cc << 53) & PGSTE_HR_BIT;
 	}
+	/* Reflect guest's logical view, not physical */
+	cc |= (pgste_val(old) & (PGSTE_GR_BIT | PGSTE_GC_BIT)) >> 49;
+	/* Changing the guest storage key is considered a change of the page */
+	if ((pgste_val(new) ^ pgste_val(old)) & PGSTE_GR_BIT)
+		pgste_val(new) |= PGSTE_UC_BIT;
 
+	pgste_set_unlock(ptep, new);
+	pte_unmap_unlock(ptep, ptl);
+	return 0;
+}
+EXPORT_SYMBOL(reset_guest_reference_bit);
+
+int get_guest_storage_key(struct mm_struct *mm, unsigned long addr,
+			  unsigned char *key)
+{
+	spinlock_t *ptl;
+	pgste_t pgste;
+	pte_t *ptep;
+
+	ptep = get_locked_pte(mm, addr, &ptl);
+	if (unlikely(!ptep))
+		return -EFAULT;
+
+	pgste = pgste_get_lock(ptep);
+	*key = (pgste_val(pgste) & (PGSTE_ACC_BITS | PGSTE_FP_BIT)) >> 56;
+	if (!(pte_val(*ptep) & _PAGE_INVALID))
+		*key = page_get_storage_key(pte_val(*ptep) & PAGE_MASK);
+	/* Reflect guest's logical view, not physical */
+	*key |= (pgste_val(pgste) & (PGSTE_GR_BIT | PGSTE_GC_BIT)) >> 48;
 	pgste_set_unlock(ptep, pgste);
 	pte_unmap_unlock(ptep, ptl);
-	up_read(&mm->mmap_sem);
-	return key;
+	return 0;
 }
 EXPORT_SYMBOL(get_guest_storage_key);
 #endif

drivers/s390/char/sclp_early.c

@@ -46,7 +46,8 @@ struct read_info_sccb {
 	u64	rnmax2;			/* 104-111 */
 	u8	_pad_112[116 - 112];	/* 112-115 */
 	u8	fac116;			/* 116 */
-	u8	_pad_117[119 - 117];	/* 117-118 */
+	u8	fac117;			/* 117 */
+	u8	_pad_118;		/* 118 */
 	u8	fac119;			/* 119 */
 	u16	hcpua;			/* 120-121 */
 	u8	_pad_122[124 - 122];	/* 122-123 */
@@ -114,7 +115,12 @@ static void __init sclp_facilities_detect(struct read_info_sccb *sccb)
 	sclp.facilities = sccb->facilities;
 	sclp.has_sprp = !!(sccb->fac84 & 0x02);
 	sclp.has_core_type = !!(sccb->fac84 & 0x01);
+	sclp.has_gsls = !!(sccb->fac85 & 0x80);
+	sclp.has_64bscao = !!(sccb->fac116 & 0x80);
+	sclp.has_cmma = !!(sccb->fac116 & 0x40);
 	sclp.has_esca = !!(sccb->fac116 & 0x08);
+	sclp.has_pfmfi = !!(sccb->fac117 & 0x40);
+	sclp.has_ibs = !!(sccb->fac117 & 0x20);
 	sclp.has_hvs = !!(sccb->fac119 & 0x80);
 	if (sccb->fac85 & 0x02)
 		S390_lowcore.machine_flags |= MACHINE_FLAG_ESOP;
@@ -145,6 +151,10 @@ static void __init sclp_facilities_detect(struct read_info_sccb *sccb)
 		sclp.has_siif = cpue->siif;
 		sclp.has_sigpif = cpue->sigpif;
 		sclp.has_sief2 = cpue->sief2;
+		sclp.has_gpere = cpue->gpere;
+		sclp.has_ib = cpue->ib;
+		sclp.has_cei = cpue->cei;
+		sclp.has_skey = cpue->skey;
 		break;
 	}
 

drivers/s390/char/sclp_ocf.c

@@ -26,7 +26,7 @@
 #define OCF_LENGTH_CPC_NAME 8UL
 
 static char hmc_network[OCF_LENGTH_HMC_NETWORK + 1];
-static char cpc_name[OCF_LENGTH_CPC_NAME + 1];
+static char cpc_name[OCF_LENGTH_CPC_NAME]; /* in EBCDIC */
 
 static DEFINE_SPINLOCK(sclp_ocf_lock);
 static struct work_struct sclp_ocf_change_work;
@@ -72,9 +72,8 @@ static void sclp_ocf_handler(struct evbuf_header *evbuf)
 	}
 	if (cpc) {
 		size = min(OCF_LENGTH_CPC_NAME, (size_t) cpc->length);
+		memset(cpc_name, 0, OCF_LENGTH_CPC_NAME);
 		memcpy(cpc_name, cpc + 1, size);
-		EBCASC(cpc_name, size);
-		cpc_name[size] = 0;
 	}
 	spin_unlock(&sclp_ocf_lock);
 	schedule_work(&sclp_ocf_change_work);
@@ -85,15 +84,23 @@ static struct sclp_register sclp_ocf_event = {
 	.receiver_fn = sclp_ocf_handler,
 };
 
+void sclp_ocf_cpc_name_copy(char *dst)
+{
+	spin_lock_irq(&sclp_ocf_lock);
+	memcpy(dst, cpc_name, OCF_LENGTH_CPC_NAME);
+	spin_unlock_irq(&sclp_ocf_lock);
+}
+EXPORT_SYMBOL(sclp_ocf_cpc_name_copy);
+
 static ssize_t cpc_name_show(struct kobject *kobj,
 			     struct kobj_attribute *attr, char *page)
 {
-	int rc;
+	char name[OCF_LENGTH_CPC_NAME + 1];
 
-	spin_lock_irq(&sclp_ocf_lock);
-	rc = snprintf(page, PAGE_SIZE, "%s\n", cpc_name);
-	spin_unlock_irq(&sclp_ocf_lock);
-	return rc;
+	sclp_ocf_cpc_name_copy(name);
+	name[OCF_LENGTH_CPC_NAME] = 0;
+	EBCASC(name, OCF_LENGTH_CPC_NAME);
+	return snprintf(page, PAGE_SIZE, "%s\n", name);
 }
 
 static struct kobj_attribute cpc_name_attr =