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

Pull s390 updates from Martin Schwidefsky:
 "There are a couple of new things for s390 with this merge request:

   - a new scheduling domain "drawer" is added to reflect the unusual
     topology found on z13 machines.  Performance tests showed up to 8
     percent gain with the additional domain.

   - the new crc-32 checksum crypto module uses the vector-galois-field
     multiply and sum SIMD instruction to speed up crc-32 and crc-32c.

   - proper __ro_after_init support, this requires RO_AFTER_INIT_DATA in
     the generic vmlinux.lds linker script definitions.

   - kcov instrumentation support.  A prerequisite for that is the
     inline assembly basic block cleanup, which is the reason for the
     net/iucv/iucv.c change.

   - support for 2GB pages is added to the hugetlbfs backend.

  Then there are two removals:

   - the oprofile hardware sampling support is dead code and is removed.
     The oprofile user space uses the perf interface nowadays.

   - the ETR clock synchronization is removed, this has been superseeded
     be the STP clock synchronization.  And it always has been
     "interesting" code..

  And the usual bug fixes and cleanups"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (82 commits)
  s390/pci: Delete an unnecessary check before the function call "pci_dev_put"
  s390/smp: clean up a condition
  s390/cio/chp : Remove deprecated create_singlethread_workqueue
  s390/chsc: improve channel path descriptor determination
  s390/chsc: sanitize fmt check for chp_desc determination
  s390/cio: make fmt1 channel path descriptor optional
  s390/chsc: fix ioctl CHSC_INFO_CU command
  s390/cio/device_ops: fix kernel doc
  s390/cio: allow to reset channel measurement block
  s390/console: Make preferred console handling more consistent
  s390/mm: fix gmap tlb flush issues
  s390/mm: add support for 2GB hugepages
  s390: have unique symbol for __switch_to address
  s390/cpuinfo: show maximum thread id
  s390/ptrace: clarify bits in the per_struct
  s390: stack address vs thread_info
  s390: remove pointless load within __switch_to
  s390: enable kcov support
  s390/cpumf: use basic block for ecctr inline assembly
  s390/hypfs: use basic block for diag inline assembly
  ...

87 files changed, 2550 insertions, 3830 deletions

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index a8c259059adf..9e607bf2d640 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -72,6 +72,7 @@ config S390
 	select ARCH_HAS_DEVMEM_IS_ALLOWED
 	select ARCH_HAS_ELF_RANDOMIZE
 	select ARCH_HAS_GCOV_PROFILE_ALL
+	select ARCH_HAS_KCOV
 	select ARCH_HAS_SG_CHAIN
 	select ARCH_HAVE_NMI_SAFE_CMPXCHG
 	select ARCH_INLINE_READ_LOCK
@@ -163,6 +164,7 @@ config S390
 	select NO_BOOTMEM
 	select OLD_SIGACTION
 	select OLD_SIGSUSPEND3
+	select SPARSE_IRQ
 	select SYSCTL_EXCEPTION_TRACE
 	select TTY
 	select VIRT_CPU_ACCOUNTING
@@ -477,6 +479,9 @@ config SCHED_MC
 config SCHED_BOOK
 	def_bool n
 
+config SCHED_DRAWER
+	def_bool n
+
 config SCHED_TOPOLOGY
 	def_bool y
 	prompt "Topology scheduler support"
@@ -484,6 +489,7 @@ config SCHED_TOPOLOGY
 	select SCHED_SMT
 	select SCHED_MC
 	select SCHED_BOOK
+	select SCHED_DRAWER
 	help
 	  Topology scheduler support improves the CPU scheduler's decision
 	  making when dealing with machines that have multi-threading,
@@ -605,16 +611,6 @@ config PCI_NR_FUNCTIONS
 	  This allows you to specify the maximum number of PCI functions which
 	  this kernel will support.
 
-config PCI_NR_MSI
-	int "Maximum number of MSI interrupts (64-32768)"
-	range 64 32768
-	default "256"
-	help
-	  This defines the number of virtual interrupts the kernel will
-	  provide for MSI interrupts. If you configure your system to have
-	  too few drivers will fail to allocate MSI interrupts for all
-	  PCI devices.
-
 source "drivers/pci/Kconfig"
 
 endif	# PCI
diff --git a/arch/s390/boot/compressed/Makefile b/arch/s390/boot/compressed/Makefile
index 1dd210347e12..98ec652cc332 100644
--- a/arch/s390/boot/compressed/Makefile
+++ b/arch/s390/boot/compressed/Makefile
@@ -4,6 +4,8 @@
 # create a compressed vmlinux image from the original vmlinux
 #
 
+KCOV_INSTRUMENT := n
+
 targets	:= vmlinux.lds vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2
 targets += vmlinux.bin.xz vmlinux.bin.lzma vmlinux.bin.lzo vmlinux.bin.lz4
 targets += misc.o piggy.o sizes.h head.o
diff --git a/arch/s390/configs/default_defconfig b/arch/s390/configs/default_defconfig
index d5ec71b2ed02..889ea3450210 100644
--- a/arch/s390/configs/default_defconfig
+++ b/arch/s390/configs/default_defconfig
@@ -678,6 +678,7 @@ CONFIG_CRYPTO_SHA512_S390=m
 CONFIG_CRYPTO_DES_S390=m
 CONFIG_CRYPTO_AES_S390=m
 CONFIG_CRYPTO_GHASH_S390=m
+CONFIG_CRYPTO_CRC32_S390=m
 CONFIG_ASYMMETRIC_KEY_TYPE=y
 CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=m
 CONFIG_X509_CERTIFICATE_PARSER=m
diff --git a/arch/s390/configs/gcov_defconfig b/arch/s390/configs/gcov_defconfig
index f46a35115d2d..1bcfd764910a 100644
--- a/arch/s390/configs/gcov_defconfig
+++ b/arch/s390/configs/gcov_defconfig
@@ -616,6 +616,7 @@ CONFIG_CRYPTO_SHA512_S390=m
 CONFIG_CRYPTO_DES_S390=m
 CONFIG_CRYPTO_AES_S390=m
 CONFIG_CRYPTO_GHASH_S390=m
+CONFIG_CRYPTO_CRC32_S390=m
 CONFIG_ASYMMETRIC_KEY_TYPE=y
 CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=m
 CONFIG_X509_CERTIFICATE_PARSER=m
diff --git a/arch/s390/configs/performance_defconfig b/arch/s390/configs/performance_defconfig
index ba0f2a58b8cd..13ff090139c8 100644
--- a/arch/s390/configs/performance_defconfig
+++ b/arch/s390/configs/performance_defconfig
@@ -615,6 +615,7 @@ CONFIG_CRYPTO_SHA512_S390=m
 CONFIG_CRYPTO_DES_S390=m
 CONFIG_CRYPTO_AES_S390=m
 CONFIG_CRYPTO_GHASH_S390=m
+CONFIG_CRYPTO_CRC32_S390=m
 CONFIG_ASYMMETRIC_KEY_TYPE=y
 CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=m
 CONFIG_X509_CERTIFICATE_PARSER=m
diff --git a/arch/s390/crypto/Makefile b/arch/s390/crypto/Makefile
index 7f0b7cda6259..d1033de4c4ee 100644
--- a/arch/s390/crypto/Makefile
+++ b/arch/s390/crypto/Makefile
@@ -9,3 +9,6 @@ obj-$(CONFIG_CRYPTO_DES_S390) += des_s390.o
 obj-$(CONFIG_CRYPTO_AES_S390) += aes_s390.o
 obj-$(CONFIG_S390_PRNG) += prng.o
 obj-$(CONFIG_CRYPTO_GHASH_S390) += ghash_s390.o
+obj-$(CONFIG_CRYPTO_CRC32_S390) += crc32-vx_s390.o
+
+crc32-vx_s390-y := crc32-vx.o crc32le-vx.o crc32be-vx.o
diff --git a/arch/s390/crypto/crc32-vx.c b/arch/s390/crypto/crc32-vx.c
new file mode 100644
index 000000000000..577ae1d4ae89
--- /dev/null
+++ b/arch/s390/crypto/crc32-vx.c
@@ -0,0 +1,310 @@
+/*
+ * Crypto-API module for CRC-32 algorithms implemented with the
+ * z/Architecture Vector Extension Facility.
+ *
+ * Copyright IBM Corp. 2015
+ * Author(s): Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
+ */
+#define KMSG_COMPONENT	"crc32-vx"
+#define pr_fmt(fmt)	KMSG_COMPONENT ": " fmt
+
+#include <linux/module.h>
+#include <linux/cpufeature.h>
+#include <linux/crc32.h>
+#include <crypto/internal/hash.h>
+#include <asm/fpu/api.h>
+
+
+#define CRC32_BLOCK_SIZE	1
+#define CRC32_DIGEST_SIZE	4
+
+#define VX_MIN_LEN		64
+#define VX_ALIGNMENT		16L
+#define VX_ALIGN_MASK		(VX_ALIGNMENT - 1)
+
+struct crc_ctx {
+	u32 key;
+};
+
+struct crc_desc_ctx {
+	u32 crc;
+};
+
+/* Prototypes for functions in assembly files */
+u32 crc32_le_vgfm_16(u32 crc, unsigned char const *buf, size_t size);
+u32 crc32_be_vgfm_16(u32 crc, unsigned char const *buf, size_t size);
+u32 crc32c_le_vgfm_16(u32 crc, unsigned char const *buf, size_t size);
+
+/*
+ * DEFINE_CRC32_VX() - Define a CRC-32 function using the vector extension
+ *
+ * Creates a function to perform a particular CRC-32 computation. Depending
+ * on the message buffer, the hardware-accelerated or software implementation
+ * is used.   Note that the message buffer is aligned to improve fetch
+ * operations of VECTOR LOAD MULTIPLE instructions.
+ *
+ */
+#define DEFINE_CRC32_VX(___fname, ___crc32_vx, ___crc32_sw)		    \
+	static u32 __pure ___fname(u32 crc,				    \
+				unsigned char const *data, size_t datalen)  \
+	{								    \
+		struct kernel_fpu vxstate;				    \
+		unsigned long prealign, aligned, remaining;		    \
+									    \
+		if ((unsigned long)data & VX_ALIGN_MASK) {		    \
+			prealign = VX_ALIGNMENT -			    \
+				  ((unsigned long)data & VX_ALIGN_MASK);    \
+			datalen -= prealign;				    \
+			crc = ___crc32_sw(crc, data, prealign);		    \
+			data = (void *)((unsigned long)data + prealign);    \
+		}							    \
+									    \
+		if (datalen < VX_MIN_LEN)				    \
+			return ___crc32_sw(crc, data, datalen);		    \
+									    \
+		aligned = datalen & ~VX_ALIGN_MASK;			    \
+		remaining = datalen & VX_ALIGN_MASK;			    \
+									    \
+		kernel_fpu_begin(&vxstate, KERNEL_VXR_LOW);		    \
+		crc = ___crc32_vx(crc, data, aligned);			    \
+		kernel_fpu_end(&vxstate);				    \
+									    \
+		if (remaining)						    \
+			crc = ___crc32_sw(crc, data + aligned, remaining);  \
+									    \
+		return crc;						    \
+	}
+
+DEFINE_CRC32_VX(crc32_le_vx, crc32_le_vgfm_16, crc32_le)
+DEFINE_CRC32_VX(crc32_be_vx, crc32_be_vgfm_16, crc32_be)
+DEFINE_CRC32_VX(crc32c_le_vx, crc32c_le_vgfm_16, __crc32c_le)
+
+
+static int crc32_vx_cra_init_zero(struct crypto_tfm *tfm)
+{
+	struct crc_ctx *mctx = crypto_tfm_ctx(tfm);
+
+	mctx->key = 0;
+	return 0;
+}
+
+static int crc32_vx_cra_init_invert(struct crypto_tfm *tfm)
+{
+	struct crc_ctx *mctx = crypto_tfm_ctx(tfm);
+
+	mctx->key = ~0;
+	return 0;
+}
+
+static int crc32_vx_init(struct shash_desc *desc)
+{
+	struct crc_ctx *mctx = crypto_shash_ctx(desc->tfm);
+	struct crc_desc_ctx *ctx = shash_desc_ctx(desc);
+
+	ctx->crc = mctx->key;
+	return 0;
+}
+
+static int crc32_vx_setkey(struct crypto_shash *tfm, const u8 *newkey,
+			   unsigned int newkeylen)
+{
+	struct crc_ctx *mctx = crypto_shash_ctx(tfm);
+
+	if (newkeylen != sizeof(mctx->key)) {
+		crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
+		return -EINVAL;
+	}
+	mctx->key = le32_to_cpu(*(__le32 *)newkey);
+	return 0;
+}
+
+static int crc32be_vx_setkey(struct crypto_shash *tfm, const u8 *newkey,
+			     unsigned int newkeylen)
+{
+	struct crc_ctx *mctx = crypto_shash_ctx(tfm);
+
+	if (newkeylen != sizeof(mctx->key)) {
+		crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
+		return -EINVAL;
+	}
+	mctx->key = be32_to_cpu(*(__be32 *)newkey);
+	return 0;
+}
+
+static int crc32le_vx_final(struct shash_desc *desc, u8 *out)
+{
+	struct crc_desc_ctx *ctx = shash_desc_ctx(desc);
+
+	*(__le32 *)out = cpu_to_le32p(&ctx->crc);
+	return 0;
+}
+
+static int crc32be_vx_final(struct shash_desc *desc, u8 *out)
+{
+	struct crc_desc_ctx *ctx = shash_desc_ctx(desc);
+
+	*(__be32 *)out = cpu_to_be32p(&ctx->crc);
+	return 0;
+}
+
+static int crc32c_vx_final(struct shash_desc *desc, u8 *out)
+{
+	struct crc_desc_ctx *ctx = shash_desc_ctx(desc);
+
+	/*
+	 * Perform a final XOR with 0xFFFFFFFF to be in sync
+	 * with the generic crc32c shash implementation.
+	 */
+	*(__le32 *)out = ~cpu_to_le32p(&ctx->crc);
+	return 0;
+}
+
+static int __crc32le_vx_finup(u32 *crc, const u8 *data, unsigned int len,
+			      u8 *out)
+{
+	*(__le32 *)out = cpu_to_le32(crc32_le_vx(*crc, data, len));
+	return 0;
+}
+
+static int __crc32be_vx_finup(u32 *crc, const u8 *data, unsigned int len,
+			      u8 *out)
+{
+	*(__be32 *)out = cpu_to_be32(crc32_be_vx(*crc, data, len));
+	return 0;
+}
+
+static int __crc32c_vx_finup(u32 *crc, const u8 *data, unsigned int len,
+			     u8 *out)
+{
+	/*
+	 * Perform a final XOR with 0xFFFFFFFF to be in sync
+	 * with the generic crc32c shash implementation.
+	 */
+	*(__le32 *)out = ~cpu_to_le32(crc32c_le_vx(*crc, data, len));
+	return 0;
+}
+
+
+#define CRC32_VX_FINUP(alg, func)					      \
+	static int alg ## _vx_finup(struct shash_desc *desc, const u8 *data,  \
+				   unsigned int datalen, u8 *out)	      \
+	{								      \
+		return __ ## alg ## _vx_finup(shash_desc_ctx(desc),	      \
+					      data, datalen, out);	      \
+	}
+
+CRC32_VX_FINUP(crc32le, crc32_le_vx)
+CRC32_VX_FINUP(crc32be, crc32_be_vx)
+CRC32_VX_FINUP(crc32c, crc32c_le_vx)
+
+#define CRC32_VX_DIGEST(alg, func)					      \
+	static int alg ## _vx_digest(struct shash_desc *desc, const u8 *data, \
+				     unsigned int len, u8 *out)		      \
+	{								      \
+		return __ ## alg ## _vx_finup(crypto_shash_ctx(desc->tfm),    \
+					      data, len, out);		      \
+	}
+
+CRC32_VX_DIGEST(crc32le, crc32_le_vx)
+CRC32_VX_DIGEST(crc32be, crc32_be_vx)
+CRC32_VX_DIGEST(crc32c, crc32c_le_vx)
+
+#define CRC32_VX_UPDATE(alg, func)					      \
+	static int alg ## _vx_update(struct shash_desc *desc, const u8 *data, \
+				     unsigned int datalen)		      \
+	{								      \
+		struct crc_desc_ctx *ctx = shash_desc_ctx(desc);	      \
+		ctx->crc = func(ctx->crc, data, datalen);		      \
+		return 0;						      \
+	}
+
+CRC32_VX_UPDATE(crc32le, crc32_le_vx)
+CRC32_VX_UPDATE(crc32be, crc32_be_vx)
+CRC32_VX_UPDATE(crc32c, crc32c_le_vx)
+
+
+static struct shash_alg crc32_vx_algs[] = {
+	/* CRC-32 LE */
+	{
+		.init		=	crc32_vx_init,
+		.setkey		=	crc32_vx_setkey,
+		.update		=	crc32le_vx_update,
+		.final		=	crc32le_vx_final,
+		.finup		=	crc32le_vx_finup,
+		.digest		=	crc32le_vx_digest,
+		.descsize	=	sizeof(struct crc_desc_ctx),
+		.digestsize	=	CRC32_DIGEST_SIZE,
+		.base		=	{
+			.cra_name	 = "crc32",
+			.cra_driver_name = "crc32-vx",
+			.cra_priority	 = 200,
+			.cra_blocksize	 = CRC32_BLOCK_SIZE,
+			.cra_ctxsize	 = sizeof(struct crc_ctx),
+			.cra_module	 = THIS_MODULE,
+			.cra_init	 = crc32_vx_cra_init_zero,
+		},
+	},
+	/* CRC-32 BE */
+	{
+		.init		=	crc32_vx_init,
+		.setkey		=	crc32be_vx_setkey,
+		.update		=	crc32be_vx_update,
+		.final		=	crc32be_vx_final,
+		.finup		=	crc32be_vx_finup,
+		.digest		=	crc32be_vx_digest,
+		.descsize	=	sizeof(struct crc_desc_ctx),
+		.digestsize	=	CRC32_DIGEST_SIZE,
+		.base		=	{
+			.cra_name	 = "crc32be",
+			.cra_driver_name = "crc32be-vx",
+			.cra_priority	 = 200,
+			.cra_blocksize	 = CRC32_BLOCK_SIZE,
+			.cra_ctxsize	 = sizeof(struct crc