Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux

Pull m ore s390 updates from Martin Schwidefsky:
 "Over 95% of the changes in this pull request are related to the zcrypt
  driver. There are five improvements for zcrypt: the ID for the CEX6
  cards is added, workload balancing and multi-domain support are
  introduced, the debug logs are overhauled and a set of tracepoints is
  added.

  Then there are several patches in regard to inline assemblies. One
  compile fix and several missing memory clobbers. As far as we can tell
  the omitted memory clobbers have not caused any breakage.

  A small change to the PCI arch code, the machine can tells us how big
  the function measurement blocks are. The PCI function measurement will
  be disabled for a device if the queried length is larger than the
  allocated size for these blocks.

  And two more patches to correct five printk messages.

  That is it for s390 in regard to the 4.10 merge window. Happy holidays"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (23 commits)
  s390/pci: query fmb length
  s390/zcrypt: add missing memory clobber to ap_qci inline assembly
  s390/extmem: add missing memory clobber to dcss_set_subcodes
  s390/nmi: fix inline assembly constraints
  s390/lib: add missing memory barriers to string inline assemblies
  s390/cpumf: fix qsi inline assembly
  s390/setup: reword printk messages
  s390/dasd: fix typos in DASD error messages
  s390: fix compile error with memmove_early() inline assembly
  s390/zcrypt: tracepoint definitions for zcrypt device driver.
  s390/zcrypt: Rework debug feature invocations.
  s390/zcrypt: Improved invalid domain response handling.
  s390/zcrypt: Fix ap_max_domain_id for older machine types
  s390/zcrypt: Correct function bits for CEX2x and CEX3x cards.
  s390/zcrypt: Fixed attrition of AP adapters and domains
  s390/zcrypt: Introduce new zcrypt device status API
  s390/zcrypt: add multi domain support
  s390/zcrypt: Introduce workload balancing
  s390/zcrypt: get rid of ap_poll_requests
  s390/zcrypt: header for the AP inline assmblies
  ...

13 files changed, 191 insertions, 32 deletions

diff --git a/arch/s390/include/asm/cpu_mf.h b/arch/s390/include/asm/cpu_mf.h
index b69d8bc231a5..428c41239a49 100644
--- a/arch/s390/include/asm/cpu_mf.h
+++ b/arch/s390/include/asm/cpu_mf.h
@@ -213,18 +213,14 @@ static inline int stcctm5(u64 num, u64 *val)
 /* Query sampling information */
 static inline int qsi(struct hws_qsi_info_block *info)
 {
-	int cc;
-	cc = 1;
+	int cc = 1;
 
 	asm volatile(
-		"0:	.insn	s,0xb2860000,0(%1)\n"
+		"0:	.insn	s,0xb2860000,%1\n"
 		"1:	lhi	%0,0\n"
 		"2:\n"
 		EX_TABLE(0b, 2b) EX_TABLE(1b, 2b)
-		: "=d" (cc), "+a" (info)
-		: "m" (*info)
-		: "cc", "memory");
-
+		: "+d" (cc), "+Q" (*info));
 	return cc ? -EINVAL : 0;
 }
 
diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h
index 6611f798d2be..4e3186649578 100644
--- a/arch/s390/include/asm/pci.h
+++ b/arch/s390/include/asm/pci.h
@@ -133,6 +133,7 @@ struct zpci_dev {
 	/* Function measurement block */
 	struct zpci_fmb *fmb;
 	u16		fmb_update;	/* update interval */
+	u16		fmb_length;
 	/* software counters */
 	atomic64_t allocated_pages;
 	atomic64_t mapped_pages;
diff --git a/arch/s390/include/asm/pci_clp.h b/arch/s390/include/asm/pci_clp.h
index c232ef9711f5..d6f1b1d94352 100644
--- a/arch/s390/include/asm/pci_clp.h
+++ b/arch/s390/include/asm/pci_clp.h
@@ -87,7 +87,8 @@ struct clp_rsp_query_pci {
 	u16 pchid;
 	u32 bar[PCI_BAR_COUNT];
 	u8 pfip[CLP_PFIP_NR_SEGMENTS];	/* pci function internal path */
-	u32			: 24;
+	u32			: 16;
+	u8 fmb_len;
 	u8 pft;				/* pci function type */
 	u64 sdma;			/* start dma as */
 	u64 edma;			/* end dma as */
diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h
index 15a3c005c274..e5f5c7074f2c 100644
--- a/arch/s390/include/asm/string.h
+++ b/arch/s390/include/asm/string.h
@@ -62,7 +62,7 @@ static inline void *memchr(const void * s, int c, size_t n)
 		"	jl	1f\n"
 		"	la	%0,0\n"
 		"1:"
-		: "+a" (ret), "+&a" (s) : "d" (r0) : "cc");
+		: "+a" (ret), "+&a" (s) : "d" (r0) : "cc", "memory");
 	return (void *) ret;
 }
 
@@ -74,7 +74,7 @@ static inline void *memscan(void *s, int c, size_t n)
 	asm volatile(
 		"0:	srst	%0,%1\n"
 		"	jo	0b\n"
-		: "+a" (ret), "+&a" (s) : "d" (r0) : "cc");
+		: "+a" (ret), "+&a" (s) : "d" (r0) : "cc", "memory");
 	return (void *) ret;
 }
 
@@ -115,7 +115,7 @@ static inline size_t strlen(const char *s)
 	asm volatile(
 		"0:	srst	%0,%1\n"
 		"	jo	0b"
-		: "+d" (r0), "+a" (tmp) :  : "cc");
+		: "+d" (r0), "+a" (tmp) :  : "cc", "memory");
 	return r0 - (unsigned long) s;
 }
 
@@ -128,7 +128,7 @@ static inline size_t strnlen(const char * s, size_t n)
 	asm volatile(
 		"0:	srst	%0,%1\n"
 		"	jo	0b"
-		: "+a" (end), "+a" (tmp) : "d" (r0)  : "cc");
+		: "+a" (end), "+a" (tmp) : "d" (r0)  : "cc", "memory");
 	return end - s;
 }
 #else /* IN_ARCH_STRING_C */
diff --git a/arch/s390/include/asm/trace/zcrypt.h b/arch/s390/include/asm/trace/zcrypt.h
new file mode 100644
index 000000000000..adcb77fafa9d
--- /dev/null
+++ b/arch/s390/include/asm/trace/zcrypt.h
@@ -0,0 +1,122 @@
+/*
+ * Tracepoint definitions for the s390 zcrypt device driver
+ *
+ * Copyright IBM Corp. 2016
+ * Author(s): Harald Freudenberger <freude@de.ibm.com>
+ *
+ * Currently there are two tracepoint events defined here.
+ * An s390_zcrypt_req request event occurs as soon as the request is
+ * recognized by the zcrypt ioctl function. This event may act as some kind
+ * of request-processing-starts-now indication.
+ * As late as possible within the zcrypt ioctl function there occurs the
+ * s390_zcrypt_rep event which may act as the point in time where the
+ * request has been processed by the kernel and the result is about to be
+ * transferred back to userspace.
+ * The glue which binds together request and reply event is the ptr
+ * parameter, which is the local buffer address where the request from
+ * userspace has been stored by the ioctl function.
+ *
+ * The main purpose of this zcrypt tracepoint api is to get some data for
+ * performance measurements together with information about on which card
+ * and queue the request has been processed. It is not an ffdc interface as
+ * there is already code in the zcrypt device driver to serve the s390
+ * debug feature interface.
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM s390
+
+#if !defined(_TRACE_S390_ZCRYPT_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_S390_ZCRYPT_H
+
+#include <linux/tracepoint.h>
+
+#define TP_ICARSAMODEXPO  0x0001
+#define TP_ICARSACRT	  0x0002
+#define TB_ZSECSENDCPRB   0x0003
+#define TP_ZSENDEP11CPRB  0x0004
+#define TP_HWRNGCPRB	  0x0005
+
+#define show_zcrypt_tp_type(type)				\
+	__print_symbolic(type,					\
+			 { TP_ICARSAMODEXPO, "ICARSAMODEXPO" }, \
+			 { TP_ICARSACRT, "ICARSACRT" },		\
+			 { TB_ZSECSENDCPRB, "ZSECSENDCPRB" },	\
+			 { TP_ZSENDEP11CPRB, "ZSENDEP11CPRB" }, \
+			 { TP_HWRNGCPRB, "HWRNGCPRB" })
+
+/**
+ * trace_s390_zcrypt_req - zcrypt request tracepoint function
+ * @ptr:  Address of the local buffer where the request from userspace
+ *	  is stored. Can be used as a unique id to relate together
+ *	  request and reply.
+ * @type: One of the TP_ defines above.
+ *
+ * Called when a request from userspace is recognised within the ioctl
+ * function of the zcrypt device driver and may act as an entry
+ * timestamp.
+ */
+TRACE_EVENT(s390_zcrypt_req,
+	    TP_PROTO(void *ptr, u32 type),
+	    TP_ARGS(ptr, type),
+	    TP_STRUCT__entry(
+		    __field(void *, ptr)
+		    __field(u32, type)),
+	    TP_fast_assign(
+		    __entry->ptr = ptr;
+		    __entry->type = type;),
+	    TP_printk("ptr=%p type=%s",
+		      __entry->ptr,
+		      show_zcrypt_tp_type(__entry->type))
+);
+
+/**
+ * trace_s390_zcrypt_rep - zcrypt reply tracepoint function
+ * @ptr:  Address of the local buffer where the request from userspace
+ *	  is stored. Can be used as a unique id to match together
+ *	  request and reply.
+ * @fc:   Function code.
+ * @rc:   The bare returncode as returned by the device driver ioctl
+ *	  function.
+ * @dev:  The adapter nr where this request was actually processed.
+ * @dom:  Domain id of the device where this request was processed.
+ *
+ * Called upon recognising the reply from the crypto adapter. This
+ * message may act as the exit timestamp for the request but also
+ * carries some info about on which adapter the request was processed
+ * and the returncode from the device driver.
+ */
+TRACE_EVENT(s390_zcrypt_rep,
+	    TP_PROTO(void *ptr, u32 fc, u32 rc, u16 dev, u16 dom),
+	    TP_ARGS(ptr, fc, rc, dev, dom),
+	    TP_STRUCT__entry(
+		    __field(void *, ptr)
+		    __field(u32, fc)
+		    __field(u32, rc)
+		    __field(u16, device)
+		    __field(u16, domain)),
+	    TP_fast_assign(
+		    __entry->ptr = ptr;
+		    __entry->fc = fc;
+		    __entry->rc = rc;
+		    __entry->device = dev;
+		    __entry->domain = dom;),
+	    TP_printk("ptr=%p fc=0x%04x rc=%d dev=0x%02hx domain=0x%04hx",
+		      __entry->ptr,
+		      (unsigned int) __entry->fc,
+		      (int) __entry->rc,
+		      (unsigned short) __entry->device,
+		      (unsigned short) __entry->domain)
+);
+
+#endif /* _TRACE_S390_ZCRYPT_H */
+
+/* This part must be outside protection */
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+
+#define TRACE_INCLUDE_PATH asm/trace
+#define TRACE_INCLUDE_FILE zcrypt
+
+#include <trace/define_trace.h>
diff --git a/arch/s390/include/uapi/asm/zcrypt.h b/arch/s390/include/uapi/asm/zcrypt.h
index f2b18eacaca8..a777f87ef889 100644
--- a/arch/s390/include/uapi/asm/zcrypt.h
+++ b/arch/s390/include/uapi/asm/zcrypt.h
@@ -215,6 +215,42 @@ struct ep11_urb {
 	uint64_t		resp;
 } __attribute__((packed));
 
+/**
+ * struct zcrypt_device_status
+ * @hwtype:		raw hardware type
+ * @qid:		6 bit device index, 8 bit domain
+ * @functions:		AP device function bit field 'abcdef'
+ *			a, b, c = reserved
+ *			d = CCA coprocessor
+ *			e = Accelerator
+ *			f = EP11 coprocessor
+ * @online		online status
+ * @reserved		reserved
+ */
+struct zcrypt_device_status {
+	unsigned int hwtype:8;
+	unsigned int qid:14;
+	unsigned int online:1;
+	unsigned int functions:6;
+	unsigned int reserved:3;
+};
+
+#define MAX_ZDEV_CARDIDS 64
+#define MAX_ZDEV_DOMAINS 256
+
+/**
+ * Maximum number of zcrypt devices
+ */
+#define MAX_ZDEV_ENTRIES (MAX_ZDEV_CARDIDS * MAX_ZDEV_DOMAINS)
+
+/**
+ * zcrypt_device_matrix
+ * Device matrix of all zcrypt devices
+ */
+struct zcrypt_device_matrix {
+	struct zcrypt_device_status device[MAX_ZDEV_ENTRIES];
+};
+
 #define AUTOSELECT ((unsigned int)0xFFFFFFFF)
 
 #define ZCRYPT_IOCTL_MAGIC 'z'
@@ -321,6 +357,7 @@ struct ep11_urb {
 #define ICARSACRT	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x06, 0)
 #define ZSECSENDCPRB	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x81, 0)
 #define ZSENDEP11CPRB	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x04, 0)
+#define ZDEVICESTATUS	_IOC(_IOC_READ|_IOC_WRITE, ZCRYPT_IOCTL_MAGIC, 0x4f, 0)
 
 /* New status calls */
 #define Z90STAT_TOTALCOUNT	_IOR(ZCRYPT_IOCTL_MAGIC, 0x40, int)
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index d038c8cea6cb..324f1c147a41 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -417,7 +417,7 @@ static __init void memmove_early(void *dst, const void *src, size_t n)
 		"	brctg	%[n],0b\n"
 		"1:\n"
 		: [addr] "=&d" (addr),
-		  [psw_pgm_addr] "=&Q" (S390_lowcore.program_new_psw.addr),
+		  [psw_pgm_addr] "=Q" (S390_lowcore.program_new_psw.addr),
 		  [dst] "+&a" (dst), [src] "+&a" (src),  [n] "+d" (n)
 		: [incr] "d" (incr)
 		: "cc", "memory");
diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c
index 9a32f7419d78..9862196b4b89 100644
--- a/arch/s390/kernel/nmi.c
+++ b/arch/s390/kernel/nmi.c
@@ -102,7 +102,7 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 {
 	int kill_task;
 	u64 zero;
-	void *fpt_save_area, *fpt_creg_save_area;
+	void *fpt_save_area;
 
 	kill_task = 0;
 	zero = 0;
@@ -130,7 +130,6 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 			kill_task = 1;
 	}
 	fpt_save_area = &S390_lowcore.floating_pt_save_area;
-	fpt_creg_save_area = &S390_lowcore.fpt_creg_save_area;
 	if (!mci.fc) {
 		/*
 		 * Floating point control register can't be restored.
@@ -142,11 +141,13 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 		 */
 		if (S390_lowcore.fpu_flags & KERNEL_FPC)
 			s390_handle_damage();
-		asm volatile("lfpc 0(%0)" : : "a" (&zero), "m" (zero));
+		asm volatile("lfpc %0" : : "Q" (zero));
 		if (!test_cpu_flag(CIF_FPU))
 			kill_task = 1;
-	} else
-		asm volatile("lfpc 0(%0)" : : "a" (fpt_creg_save_area));
+	} else {
+		asm volatile("lfpc %0"
+			     : : "Q" (S390_lowcore.fpt_creg_save_area));
+	}
 
 	if (!MACHINE_HAS_VX) {
 		/* Validate floating point registers */
@@ -167,7 +168,7 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 			"	ld	13,104(%0)\n"
 			"	ld	14,112(%0)\n"
 			"	ld	15,120(%0)\n"
-			: : "a" (fpt_save_area));
+			: : "a" (fpt_save_area) : "memory");
 	} else {
 		/* Validate vector registers */
 		union ctlreg0 cr0;
@@ -217,7 +218,7 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 	} else {
 		asm volatile(
 			"	lctlg	0,15,0(%0)"
-			: : "a" (&S390_lowcore.cregs_save_area));
+			: : "a" (&S390_lowcore.cregs_save_area) : "memory");
 	}
 	/*
 	 * We don't even try to validate the TOD register, since we simply
@@ -234,9 +235,9 @@ static int notrace s390_validate_registers(union mci mci, int umode)
 			: : : "0", "cc");
 	else
 		asm volatile(
-			"	l	0,0(%0)\n"
+			"	l	0,%0\n"
 			"	sckpf"
-			: : "a" (&S390_lowcore.tod_progreg_save_area)
+			: : "Q" (S390_lowcore.tod_progreg_save_area)
 			: "0", "cc");
 	/* Validate clock comparator register */
 	set_clock_comparator(S390_lowcore.clock_comparator);
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index adfac9f0a89f..865a48871ca4 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -485,7 +485,7 @@ static void __init setup_memory_end(void)
 	max_pfn = max_low_pfn = PFN_DOWN(memory_end);
 	memblock_remove(memory_end, ULONG_MAX);
 
-	pr_notice("Max memory size: %luMB\n", memory_end >> 20);
+	pr_notice("The maximum memory size is %luMB\n", memory_end >> 20);
 }
 
 static void __init setup_vmcoreinfo(void)
@@ -650,7 +650,7 @@ static void __init check_initrd(void)
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (INITRD_START && INITRD_SIZE &&
 	    !memblock_is_region_memory(INITRD_START, INITRD_SIZE)) {
-		pr_err("initrd does not fit memory.\n");
+		pr_err("The initial RAM disk does not fit into the memory\n");
 		memblock_free(INITRD_START, INITRD_SIZE);
 		initrd_start = initrd_end = 0;
 	}
diff --git a/arch/s390/lib/string.c b/arch/s390/lib/string.c
index 48352bffbc92..f71d9f655970 100644
--- a/arch/s390/lib/string.c
+++ b/arch/s390/lib/string.c
@@ -20,7 +20,7 @@ static inline char *__strend(const char *s)
 
 	asm volatile ("0: srst  %0,%1\n"
 		      "   jo    0b"
-		      : "+d" (r0), "+a" (s) :  : "cc" );
+		      : "+d" (r0), "+a" (s) :  : "cc", "memory");
 	return (char *) r0;
 }
 
@@ -31,7 +31,7 @@ static inline char *__strnend(const char *s, size_t n)
 
 	asm volatile ("0: srst  %0,%1\n"
 		      "   jo    0b"
-		      : "+d" (p), "+a" (s) : "d" (r0) : "cc" );
+		      : "+d" (p), "+a" (s) : "d" (r0) : "cc", "memory");
 	return (char *) p;
 }
 
@@ -213,7 +213,7 @@ int strcmp(const char *cs, const char *ct)
 		      "   sr   %0,%1\n"
 		      "1:"
 		      : "+d" (ret), "+d" (r0), "+a" (cs), "+a" (ct)
-		      : : "cc" );
+		      : : "cc", "memory");
 	return ret;
 }
 EXPORT_SYMBOL(strcmp);
@@ -250,7 +250,7 @@ static inline int clcle(const char *s1, unsigned long l1,
 		      "   ipm   %0\n"
 		      "   srl   %0,28"
 		      : "=&d" (cc), "+a" (r2), "+a" (r3),
-			"+a" (r4), "+a" (r5) : : "cc");
+			"+a" (r4), "+a" (r5) : : "cc", "memory");
 	return cc;
 }
 
@@ -298,7 +298,7 @@ void *memchr(const void *s, int c, size_t n)
 		      "   jl	1f\n"
 		      "   la    %0,0\n"
 		      "1:"
-		      : "+a" (ret), "+&a" (s) : "d" (r0) : "cc" );
+		      : "+a" (ret), "+&a" (s) : "d" (r0) : "cc", "memory");
 	return (void *) ret;
 }
 EXPORT_SYMBOL(memchr);
@@ -336,7 +336,7 @@ void *memscan(void *s, int c, size_t n)
 
 	asm volatile ("0: srst  %0,%1\n"
 		      "   jo    0b\n"
-		      : "+a" (ret), "+&a" (s) : "d" (r0) : "cc" );
+		      : "+a" (ret), "+&a" (s) : "d" (r0) : "cc", "memory");
 	return (void *) ret;
 }
 EXPORT_SYMBOL(memscan);
diff --git a/arch/s390/mm/extmem.c b/arch/s390/mm/extmem.c
index 02042b6b66bf..362237203144 100644
--- a/arch/s390/mm/extmem.c
+++ b/arch/s390/mm/extmem.c
@@ -122,7 +122,7 @@ dcss_set_subcodes(void)
 		"1:	la	%2,3\n"
 		"2:\n"
 		EX_TABLE(0b, 1b)
-		: "+d" (rx), "+d" (ry), "=d" (rc) : : "cc");
+		: "+d" (rx), "+d" (ry), "=d" (rc) : : "cc", "memory");
 
 	kfree(name);
 	/* Diag x'64' new subcodes are supported, set to new subcodes */
diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
index 64e1734bebb7..38e17d4d9884 100644
--- a/arch/s390/pci/pci.c
+++ b/arch/s390/pci/pci.c
@@ -180,7 +180,7 @@ int zpci_fmb_enable_device(struct zpci_dev *zdev)
 {
 	struct mod_pci_args args = { 0, 0, 0, 0 };
 
-	if (zdev->fmb)
+	if (zdev->fmb || sizeof(*zdev->fmb) < zdev->fmb_length)
 		return -EINVAL;
 
 	zdev->fmb = kmem_cache_zalloc(zdev_fmb_cache, GFP_KERNEL);
diff --git a/arch/s390/pci/pci_clp.c b/arch/s390/pci/pci_clp.c
index e3ef63b36b5a..1c3332ac1957 100644
--- a/arch/s390/pci/pci_clp.c
+++ b/arch/s390/pci/pci_clp.c
@@ -148,6 +148,7 @@ static int clp_store_query_pci_fn(struct zpci_dev *zdev,
 	zdev->pft = response->pft;
 	zdev->vfn = response->vfn;
 	zdev->uid = response->uid;
+	zdev->fmb_length = sizeof(u32) * response->fmb_len;
 
 	memcpy(zdev->pfip, response->pfip, sizeof(zdev->pfip));
 	if (response->util_str_avail) {