From cb9eff097831007afb30d64373f29d99825d0068 Mon Sep 17 00:00:00 2001
From: Patrick Ohly
Date: Thu, 12 Feb 2009 05:03:36 +0000
Subject: net: new user space API for time stamping of incoming and outgoing
 packets

User space can request hardware and/or software time stamping.
Reporting of the result(s) via a new control message is enabled
separately for each field in the message because some of the
fields may require additional computation and thus cause overhead.
User space can tell the different kinds of time stamps apart
and choose what suits its needs.

When a TX timestamp operation is requested, the TX skb will be cloned
and the clone will be time stamped (in hardware or software) and added
to the socket error queue of the skb, if the skb has a socket
associated with it.

The actual TX timestamp will reach userspace as a RX timestamp on the
cloned packet. If timestamping is requested and no timestamping is
done in the device driver (potentially this may use hardware
timestamping), it will be done in software after the device's
start_hard_xmit routine.

Signed-off-by: Patrick Ohly <patrick.ohly@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/s390/include/asm/socket.h | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/socket.h b/arch/s390/include/asm/socket.h
index c786ab623b2d..02330c50241b 100644
--- a/arch/s390/include/asm/socket.h
+++ b/arch/s390/include/asm/socket.h
@@ -62,4 +62,7 @@
 
 #define SO_MARK			36
 
+#define SO_TIMESTAMPING		37
+#define SCM_TIMESTAMPING	SO_TIMESTAMPING
+
 #endif /* _ASM_SOCKET_H */
-- 
cgit v1.2.3-55-g7522


From 563f346d04e8373739240604a51ce8529dd9f07e Mon Sep 17 00:00:00 2001
From: Herbert Xu
Date: Sun, 18 Jan 2009 20:33:33 +1100
Subject: crypto: sha-s390 - Switch to shash

This patch converts the S390 sha algorithms to the new shash interface.

With fixes by Jan Glauber.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 arch/s390/crypto/sha.h         |  6 ++--
 arch/s390/crypto/sha1_s390.c   | 40 +++++++++++----------
 arch/s390/crypto/sha256_s390.c | 40 +++++++++++----------
 arch/s390/crypto/sha512_s390.c | 80 ++++++++++++++++++++++--------------------
 arch/s390/crypto/sha_common.c  | 20 ++++++-----
 drivers/crypto/Kconfig         |  6 ++--
 6 files changed, 103 insertions(+), 89 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/crypto/sha.h b/arch/s390/crypto/sha.h
index 1ceafa571eab..f4e9dc71675f 100644
--- a/arch/s390/crypto/sha.h
+++ b/arch/s390/crypto/sha.h
@@ -29,7 +29,9 @@ struct s390_sha_ctx {
 	int func;		/* KIMD function to use */
 };
 
-void s390_sha_update(struct crypto_tfm *tfm, const u8 *data, unsigned int len);
-void s390_sha_final(struct crypto_tfm *tfm, u8 *out);
+struct shash_desc;
+
+int s390_sha_update(struct shash_desc *desc, const u8 *data, unsigned int len);
+int s390_sha_final(struct shash_desc *desc, u8 *out);
 
 #endif
diff --git a/arch/s390/crypto/sha1_s390.c b/arch/s390/crypto/sha1_s390.c
index b3cb5a89b00d..e85ba348722a 100644
--- a/arch/s390/crypto/sha1_s390.c
+++ b/arch/s390/crypto/sha1_s390.c
@@ -23,17 +23,17 @@
  * any later version.
  *
  */
+#include <crypto/internal/hash.h>
 #include <linux/init.h>
 #include <linux/module.h>
-#include <linux/crypto.h>
 #include <crypto/sha.h>
 
 #include "crypt_s390.h"
 #include "sha.h"
 
-static void sha1_init(struct crypto_tfm *tfm)
+static int sha1_init(struct shash_desc *desc)
 {
-	struct s390_sha_ctx *sctx = crypto_tfm_ctx(tfm);
+	struct s390_sha_ctx *sctx = shash_desc_ctx(desc);
 
 	sctx->state[0] = SHA1_H0;
 	sctx->state[1] = SHA1_H1;
@@ -42,34 +42,36 @@ static void sha1_init(struct crypto_tfm *tfm)
 	sctx->state[4] = SHA1_H4;
 	sctx->count = 0;
 	sctx->func = KIMD_SHA_1;
+
+	return 0;
 }
 
-static struct crypto_alg alg = {
-	.cra_name	=	"sha1",
-	.cra_driver_name=	"sha1-s390",
-	.cra_priority	=	CRYPT_S390_PRIORITY,
-	.cra_flags	=	CRYPTO_ALG_TYPE_DIGEST,
-	.cra_blocksize	=	SHA1_BLOCK_SIZE,
-	.cra_ctxsize	=	sizeof(struct s390_sha_ctx),
-	.cra_module	=	THIS_MODULE,
-	.cra_list	=	LIST_HEAD_INIT(alg.cra_list),
-	.cra_u		=	{ .digest = {
-	.dia_digestsize	=	SHA1_DIGEST_SIZE,
-	.dia_init	=	sha1_init,
-	.dia_update	=	s390_sha_update,
-	.dia_final	=	s390_sha_final } }
+static struct shash_alg alg = {
+	.digestsize	=	SHA1_DIGEST_SIZE,
+	.init		=	sha1_init,
+	.update		=	s390_sha_update,
+	.final		=	s390_sha_final,
+	.descsize	=	sizeof(struct s390_sha_ctx),
+	.base		=	{
+		.cra_name	=	"sha1",
+		.cra_driver_name=	"sha1-s390",
+		.cra_priority	=	CRYPT_S390_PRIORITY,
+		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_blocksize	=	SHA1_BLOCK_SIZE,
+		.cra_module	=	THIS_MODULE,
+	}
 };
 
 static int __init sha1_s390_init(void)
 {
 	if (!crypt_s390_func_available(KIMD_SHA_1))
 		return -EOPNOTSUPP;
-	return crypto_register_alg(&alg);
+	return crypto_register_shash(&alg);
 }
 
 static void __exit sha1_s390_fini(void)
 {
-	crypto_unregister_alg(&alg);
+	crypto_unregister_shash(&alg);
 }
 
 module_init(sha1_s390_init);
diff --git a/arch/s390/crypto/sha256_s390.c b/arch/s390/crypto/sha256_s390.c
index 19c03fb6ba7e..f9fefc569632 100644
--- a/arch/s390/crypto/sha256_s390.c
+++ b/arch/s390/crypto/sha256_s390.c
@@ -16,17 +16,17 @@
  * any later version.
  *
  */
+#include <crypto/internal/hash.h>
 #include <linux/init.h>
 #include <linux/module.h>
-#include <linux/crypto.h>
 #include <crypto/sha.h>
 
 #include "crypt_s390.h"
 #include "sha.h"
 
-static void sha256_init(struct crypto_tfm *tfm)
+static int sha256_init(struct shash_desc *desc)
 {
-	struct s390_sha_ctx *sctx = crypto_tfm_ctx(tfm);
+	struct s390_sha_ctx *sctx = shash_desc_ctx(desc);
 
 	sctx->state[0] = SHA256_H0;
 	sctx->state[1] = SHA256_H1;
@@ -38,22 +38,24 @@ static void sha256_init(struct crypto_tfm *tfm)
 	sctx->state[7] = SHA256_H7;
 	sctx->count = 0;
 	sctx->func = KIMD_SHA_256;
+
+	return 0;
 }
 
-static struct crypto_alg alg = {
-	.cra_name	=	"sha256",
-	.cra_driver_name =	"sha256-s390",
-	.cra_priority	=	CRYPT_S390_PRIORITY,
-	.cra_flags	=	CRYPTO_ALG_TYPE_DIGEST,
-	.cra_blocksize	=	SHA256_BLOCK_SIZE,
-	.cra_ctxsize	=	sizeof(struct s390_sha_ctx),
-	.cra_module	=	THIS_MODULE,
-	.cra_list	=	LIST_HEAD_INIT(alg.cra_list),
-	.cra_u		=	{ .digest = {
-	.dia_digestsize	=	SHA256_DIGEST_SIZE,
-	.dia_init	=	sha256_init,
-	.dia_update	=	s390_sha_update,
-	.dia_final	=	s390_sha_final } }
+static struct shash_alg alg = {
+	.digestsize	=	SHA256_DIGEST_SIZE,
+	.init		=	sha256_init,
+	.update		=	s390_sha_update,
+	.final		=	s390_sha_final,
+	.descsize	=	sizeof(struct s390_sha_ctx),
+	.base		=	{
+		.cra_name	=	"sha256",
+		.cra_driver_name=	"sha256-s390",
+		.cra_priority	=	CRYPT_S390_PRIORITY,
+		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_blocksize	=	SHA256_BLOCK_SIZE,
+		.cra_module	=	THIS_MODULE,
+	}
 };
 
 static int sha256_s390_init(void)
@@ -61,12 +63,12 @@ static int sha256_s390_init(void)
 	if (!crypt_s390_func_available(KIMD_SHA_256))
 		return -EOPNOTSUPP;
 
-	return crypto_register_alg(&alg);
+	return crypto_register_shash(&alg);
 }
 
 static void __exit sha256_s390_fini(void)
 {
-	crypto_unregister_alg(&alg);
+	crypto_unregister_shash(&alg);
 }
 
 module_init(sha256_s390_init);
diff --git a/arch/s390/crypto/sha512_s390.c b/arch/s390/crypto/sha512_s390.c
index 23c7861f6aeb..420acf41b727 100644
--- a/arch/s390/crypto/sha512_s390.c
+++ b/arch/s390/crypto/sha512_s390.c
@@ -12,16 +12,16 @@
  * any later version.
  *
  */
+#include <crypto/internal/hash.h>
 #include <linux/init.h>
 #include <linux/module.h>
-#include <linux/crypto.h>
 
 #include "sha.h"
 #include "crypt_s390.h"
 
-static void sha512_init(struct crypto_tfm *tfm)
+static int sha512_init(struct shash_desc *desc)
 {
-	struct s390_sha_ctx *ctx = crypto_tfm_ctx(tfm);
+	struct s390_sha_ctx *ctx = shash_desc_ctx(desc);
 
 	*(__u64 *)&ctx->state[0] = 0x6a09e667f3bcc908ULL;
 	*(__u64 *)&ctx->state[2] = 0xbb67ae8584caa73bULL;
@@ -33,29 +33,31 @@ static void sha512_init(struct crypto_tfm *tfm)
 	*(__u64 *)&ctx->state[14] = 0x5be0cd19137e2179ULL;
 	ctx->count = 0;
 	ctx->func = KIMD_SHA_512;
+
+	return 0;
 }
 
-static struct crypto_alg sha512_alg = {
-	.cra_name	=	"sha512",
-	.cra_driver_name =	"sha512-s390",
-	.cra_priority	=	CRYPT_S390_PRIORITY,
-	.cra_flags	=	CRYPTO_ALG_TYPE_DIGEST,
-	.cra_blocksize	=	SHA512_BLOCK_SIZE,
-	.cra_ctxsize	=	sizeof(struct s390_sha_ctx),
-	.cra_module	=	THIS_MODULE,
-	.cra_list	=	LIST_HEAD_INIT(sha512_alg.cra_list),
-	.cra_u		=	{ .digest = {
-	.dia_digestsize	=	SHA512_DIGEST_SIZE,
-	.dia_init	=	sha512_init,
-	.dia_update	=	s390_sha_update,
-	.dia_final	=	s390_sha_final } }
+static struct shash_alg sha512_alg = {
+	.digestsize	=	SHA512_DIGEST_SIZE,
+	.init		=	sha512_init,
+	.update		=	s390_sha_update,
+	.final		=	s390_sha_final,
+	.descsize	=	sizeof(struct s390_sha_ctx),
+	.base		=	{
+		.cra_name	=	"sha512",
+		.cra_driver_name=	"sha512-s390",
+		.cra_priority	=	CRYPT_S390_PRIORITY,
+		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_blocksize	=	SHA512_BLOCK_SIZE,
+		.cra_module	=	THIS_MODULE,
+	}
 };
 
 MODULE_ALIAS("sha512");
 
-static void sha384_init(struct crypto_tfm *tfm)
+static int sha384_init(struct shash_desc *desc)
 {
-	struct s390_sha_ctx *ctx = crypto_tfm_ctx(tfm);
+	struct s390_sha_ctx *ctx = shash_desc_ctx(desc);
 
 	*(__u64 *)&ctx->state[0] = 0xcbbb9d5dc1059ed8ULL;
 	*(__u64 *)&ctx->state[2] = 0x629a292a367cd507ULL;
@@ -67,22 +69,24 @@ static void sha384_init(struct crypto_tfm *tfm)
 	*(__u64 *)&ctx->state[14] = 0x47b5481dbefa4fa4ULL;
 	ctx->count = 0;
 	ctx->func = KIMD_SHA_512;
+
+	return 0;
 }
 
-static struct crypto_alg sha384_alg = {
-	.cra_name	=	"sha384",
-	.cra_driver_name =	"sha384-s390",
-	.cra_priority	=	CRYPT_S390_PRIORITY,
-	.cra_flags	=	CRYPTO_ALG_TYPE_DIGEST,
-	.cra_blocksize	=	SHA384_BLOCK_SIZE,
-	.cra_ctxsize	=	sizeof(struct s390_sha_ctx),
-	.cra_module	=	THIS_MODULE,
-	.cra_list	=	LIST_HEAD_INIT(sha384_alg.cra_list),
-	.cra_u		=	{ .digest = {
-	.dia_digestsize	=	SHA384_DIGEST_SIZE,
-	.dia_init	=	sha384_init,
-	.dia_update	=	s390_sha_update,
-	.dia_final	=	s390_sha_final } }
+static struct shash_alg sha384_alg = {
+	.digestsize	=	SHA384_DIGEST_SIZE,
+	.init		=	sha384_init,
+	.update		=	s390_sha_update,
+	.final		=	s390_sha_final,
+	.descsize	=	sizeof(struct s390_sha_ctx),
+	.base		=	{
+		.cra_name	=	"sha384",
+		.cra_driver_name=	"sha384-s390",
+		.cra_priority	=	CRYPT_S390_PRIORITY,
+		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_ctxsize	=	sizeof(struct s390_sha_ctx),
+		.cra_module	=	THIS_MODULE,
+	}
 };
 
 MODULE_ALIAS("sha384");
@@ -93,18 +97,18 @@ static int __init init(void)
 
 	if (!crypt_s390_func_available(KIMD_SHA_512))
 		return -EOPNOTSUPP;
-	if ((ret = crypto_register_alg(&sha512_alg)) < 0)
+	if ((ret = crypto_register_shash(&sha512_alg)) < 0)
 		goto out;
-	if ((ret = crypto_register_alg(&sha384_alg)) < 0)
-		crypto_unregister_alg(&sha512_alg);
+	if ((ret = crypto_register_shash(&sha384_alg)) < 0)
+		crypto_unregister_shash(&sha512_alg);
 out:
 	return ret;
 }
 
 static void __exit fini(void)
 {
-	crypto_unregister_alg(&sha512_alg);
-	crypto_unregister_alg(&sha384_alg);
+	crypto_unregister_shash(&sha512_alg);
+	crypto_unregister_shash(&sha384_alg);
 }
 
 module_init(init);
diff --git a/arch/s390/crypto/sha_common.c b/arch/s390/crypto/sha_common.c
index 9d6eb8c3d37e..7903ec47e6b9 100644
--- a/arch/s390/crypto/sha_common.c
+++ b/arch/s390/crypto/sha_common.c
@@ -13,14 +13,14 @@
  *
  */
 
-#include <linux/crypto.h>
+#include <crypto/internal/hash.h>
 #include "sha.h"
 #include "crypt_s390.h"
 
-void s390_sha_update(struct crypto_tfm *tfm, const u8 *data, unsigned int len)
+int s390_sha_update(struct shash_desc *desc, const u8 *data, unsigned int len)
 {
-	struct s390_sha_ctx *ctx = crypto_tfm_ctx(tfm);
-	unsigned int bsize = crypto_tfm_alg_blocksize(tfm);
+	struct s390_sha_ctx *ctx = shash_desc_ctx(desc);
+	unsigned int bsize = crypto_shash_blocksize(desc->tfm);
 	unsigned int index;
 	int ret;
 
@@ -51,13 +51,15 @@ void s390_sha_update(struct crypto_tfm *tfm, const u8 *data, unsigned int len)
 store:
 	if (len)
 		memcpy(ctx->buf + index , data, len);
+
+	return 0;
 }
 EXPORT_SYMBOL_GPL(s390_sha_update);
 
-void s390_sha_final(struct crypto_tfm *tfm, u8 *out)
+int s390_sha_final(struct shash_desc *desc, u8 *out)
 {
-	struct s390_sha_ctx *ctx = crypto_tfm_ctx(tfm);
-	unsigned int bsize = crypto_tfm_alg_blocksize(tfm);
+	struct s390_sha_ctx *ctx = shash_desc_ctx(desc);
+	unsigned int bsize = crypto_shash_blocksize(desc->tfm);
 	u64 bits;
 	unsigned int index, end, plen;
 	int ret;
@@ -87,9 +89,11 @@ void s390_sha_final(struct crypto_tfm *tfm, u8 *out)
 	BUG_ON(ret != end);
 
 	/* copy digest to out */
-	memcpy(out, ctx->state, crypto_hash_digestsize(crypto_hash_cast(tfm)));
+	memcpy(out, ctx->state, crypto_shash_digestsize(desc->tfm));
 	/* wipe context */
 	memset(ctx, 0, sizeof *ctx);
+
+	return 0;
 }
 EXPORT_SYMBOL_GPL(s390_sha_final);
 
diff --git a/drivers/crypto/Kconfig b/drivers/crypto/Kconfig
index e522144cba3a..5adbae71454f 100644
--- a/drivers/crypto/Kconfig
+++ b/drivers/crypto/Kconfig
@@ -86,7 +86,7 @@ config ZCRYPT_MONOLITHIC
 config CRYPTO_SHA1_S390
 	tristate "SHA1 digest algorithm"
 	depends on S390
-	select CRYPTO_ALGAPI
+	select CRYPTO_HASH
 	help
 	  This is the s390 hardware accelerated implementation of the
 	  SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2).
@@ -94,7 +94,7 @@ config CRYPTO_SHA1_S390
 config CRYPTO_SHA256_S390
 	tristate "SHA256 digest algorithm"
 	depends on S390
-	select CRYPTO_ALGAPI
+	select CRYPTO_HASH
 	help
 	  This is the s390 hardware accelerated implementation of the
 	  SHA256 secure hash standard (DFIPS 180-2).
@@ -105,7 +105,7 @@ config CRYPTO_SHA256_S390
 config CRYPTO_SHA512_S390
 	tristate "SHA384 and SHA512 digest algorithm"
 	depends on S390
-	select CRYPTO_ALGAPI
+	select CRYPTO_HASH
 	help
 	  This is the s390 hardware accelerated implementation of the
 	  SHA512 secure hash standard.
-- 
cgit v1.2.3-55-g7522


From 949abe574739848b1e68271fbac86c3cb4506aad Mon Sep 17 00:00:00 2001
From: Herbert Xu
Date: Sat, 21 Mar 2009 21:12:19 +0800
Subject: crypto: sha512-s390 - Add missing block size

I missed the block size when converting sha512-s390 to shash.

Tested-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 arch/s390/crypto/sha512_s390.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch/s390')

diff --git a/arch/s390/crypto/sha512_s390.c b/arch/s390/crypto/sha512_s390.c
index 420acf41b727..83192bfc8048 100644
--- a/arch/s390/crypto/sha512_s390.c
+++ b/arch/s390/crypto/sha512_s390.c
@@ -84,6 +84,7 @@ static struct shash_alg sha384_alg = {
 		.cra_driver_name=	"sha384-s390",
 		.cra_priority	=	CRYPT_S390_PRIORITY,
 		.cra_flags	=	CRYPTO_ALG_TYPE_SHASH,
+		.cra_blocksize	=	SHA384_BLOCK_SIZE,
 		.cra_ctxsize	=	sizeof(struct s390_sha_ctx),
 		.cra_module	=	THIS_MODULE,
 	}
-- 
cgit v1.2.3-55-g7522


From d0bfb940ecabf0b44fb1fd80d8d60594e569e5ec Mon Sep 17 00:00:00 2001
From: Jan Kiszka
Date: Mon, 15 Dec 2008 13:52:10 +0100
Subject: KVM: New guest debug interface

This rips out the support for KVM_DEBUG_GUEST and introduces a new IOCTL
instead: KVM_SET_GUEST_DEBUG. The IOCTL payload consists of a generic
part, controlling the "main switch" and the single-step feature. The
arch specific part adds an x86 interface for intercepting both types of
debug exceptions separately and re-injecting them when the host was not
interested. Moveover, the foundation for guest debugging via debug
registers is layed.

To signal breakpoint events properly back to userland, an arch-specific
data block is now returned along KVM_EXIT_DEBUG. For x86, the arch block
contains the PC, the debug exception, and relevant debug registers to
tell debug events properly apart.

The availability of this new interface is signaled by
KVM_CAP_SET_GUEST_DEBUG. Empty stubs for not yet supported archs are
provided.

Note that both SVM and VTX are supported, but only the latter was tested
yet. Based on the experience with all those VTX corner case, I would be
fairly surprised if SVM will work out of the box.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/ia64/include/asm/kvm.h     |  7 ++++
 arch/ia64/kvm/kvm-ia64.c        |  4 +-
 arch/powerpc/include/asm/kvm.h  |  7 ++++
 arch/powerpc/kvm/powerpc.c      |  4 +-
 arch/s390/include/asm/kvm.h     |  7 ++++
 arch/s390/kvm/kvm-s390.c        |  4 +-
 arch/x86/include/asm/kvm.h      | 18 ++++++++
 arch/x86/include/asm/kvm_host.h |  9 +---
 arch/x86/kvm/svm.c              | 50 +++++++++++++++++++++-
 arch/x86/kvm/vmx.c              | 93 ++++++++++++++++-------------------------
 arch/x86/kvm/x86.c              | 14 ++++---
 include/linux/kvm.h             | 51 +++++++++++++++-------
 include/linux/kvm_host.h        |  6 +--
 virt/kvm/kvm_main.c             |  6 +--
 14 files changed, 179 insertions(+), 101 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/ia64/include/asm/kvm.h b/arch/ia64/include/asm/kvm.h
index bfa86b6af7cd..be3fdb891214 100644
--- a/arch/ia64/include/asm/kvm.h
+++ b/arch/ia64/include/asm/kvm.h
@@ -214,4 +214,11 @@ struct kvm_sregs {
 struct kvm_fpu {
 };
 
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
 #endif
diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index 28f982045f29..de47467a0e61 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -1303,8 +1303,8 @@ int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu)
 	return -EINVAL;
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-		struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
 {
 	return -EINVAL;
 }
diff --git a/arch/powerpc/include/asm/kvm.h b/arch/powerpc/include/asm/kvm.h
index f993e4198d5c..755f1b1948c5 100644
--- a/arch/powerpc/include/asm/kvm.h
+++ b/arch/powerpc/include/asm/kvm.h
@@ -52,4 +52,11 @@ struct kvm_fpu {
 	__u64 fpr[32];
 };
 
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
 #endif /* __LINUX_KVM_POWERPC_H */
diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c
index 5f81256287f5..7c2ad4017d6a 100644
--- a/arch/powerpc/kvm/powerpc.c
+++ b/arch/powerpc/kvm/powerpc.c
@@ -240,8 +240,8 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu)
 	kvmppc_core_vcpu_put(vcpu);
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-                                    struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
 {
 	int i;
 
diff --git a/arch/s390/include/asm/kvm.h b/arch/s390/include/asm/kvm.h
index e1f54654e3ae..0b2f829f6d50 100644
--- a/arch/s390/include/asm/kvm.h
+++ b/arch/s390/include/asm/kvm.h
@@ -42,4 +42,11 @@ struct kvm_fpu {
 	__u64 fprs[16];
 };
 
+struct kvm_debug_exit_arch {
+};
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+};
+
 #endif
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 0d33893e1e89..cbfe91e10120 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -422,8 +422,8 @@ int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu,
 	return -EINVAL; /* not implemented yet */
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-				    struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
 {
 	return -EINVAL; /* not implemented yet */
 }
diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h
index 886c9402ec45..32eb96c7ca27 100644
--- a/arch/x86/include/asm/kvm.h
+++ b/arch/x86/include/asm/kvm.h
@@ -212,6 +212,24 @@ struct kvm_pit_channel_state {
 	__s64 count_load_time;
 };
 
+struct kvm_debug_exit_arch {
+	__u32 exception;
+	__u32 pad;
+	__u64 pc;
+	__u64 dr6;
+	__u64 dr7;
+};
+
+#define KVM_GUESTDBG_USE_SW_BP		0x00010000
+#define KVM_GUESTDBG_USE_HW_BP		0x00020000
+#define KVM_GUESTDBG_INJECT_DB		0x00040000
+#define KVM_GUESTDBG_INJECT_BP		0x00080000
+
+/* for KVM_SET_GUEST_DEBUG */
+struct kvm_guest_debug_arch {
+	__u64 debugreg[8];
+};
+
 struct kvm_pit_state {
 	struct kvm_pit_channel_state channels[3];
 };
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 53779309514a..c430cd580ee2 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -135,12 +135,6 @@ enum {
 
 #define KVM_NR_MEM_OBJS 40
 
-struct kvm_guest_debug {
-	int enabled;
-	unsigned long bp[4];
-	int singlestep;
-};
-
 /*
  * We don't want allocation failures within the mmu code, so we preallocate
  * enough memory for a single page fault in a cache.
@@ -448,8 +442,7 @@ struct kvm_x86_ops {
 	void (*vcpu_put)(struct kvm_vcpu *vcpu);
 
 	int (*set_guest_debug)(struct kvm_vcpu *vcpu,
-			       struct kvm_debug_guest *dbg);
-	void (*guest_debug_pre)(struct kvm_vcpu *vcpu);
+			       struct kvm_guest_debug *dbg);
 	int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata);
 	int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data);
 	u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg);
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 0fbbde54ecae..88d9062f4545 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -968,9 +968,32 @@ static void svm_set_segment(struct kvm_vcpu *vcpu,
 
 }
 
-static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
+static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg)
 {
-	return -EOPNOTSUPP;
+	int old_debug = vcpu->guest_debug;
+	struct vcpu_svm *svm = to_svm(vcpu);
+
+	vcpu->guest_debug = dbg->control;
+
+	svm->vmcb->control.intercept_exceptions &=
+		~((1 << DB_VECTOR) | (1 << BP_VECTOR));
+	if (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) {
+		if (vcpu->guest_debug &
+		    (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))
+			svm->vmcb->control.intercept_exceptions |=
+				1 << DB_VECTOR;
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP)
+			svm->vmcb->control.intercept_exceptions |=
+				1 << BP_VECTOR;
+	} else
+		vcpu->guest_debug = 0;
+
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
+		svm->vmcb->save.rflags |= X86_EFLAGS_TF | X86_EFLAGS_RF;
+	else if (old_debug & KVM_GUESTDBG_SINGLESTEP)
+		svm->vmcb->save.rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
+
+	return 0;
 }
 
 static int svm_get_irq(struct kvm_vcpu *vcpu)
@@ -1094,6 +1117,27 @@ static int pf_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 	return kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code);
 }
 
+static int db_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	if (!(svm->vcpu.guest_debug &
+	      (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))) {
+		kvm_queue_exception(&svm->vcpu, DB_VECTOR);
+		return 1;
+	}
+	kvm_run->exit_reason = KVM_EXIT_DEBUG;
+	kvm_run->debug.arch.pc = svm->vmcb->save.cs.base + svm->vmcb->save.rip;
+	kvm_run->debug.arch.exception = DB_VECTOR;
+	return 0;
+}
+
+static int bp_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
+{
+	kvm_run->exit_reason = KVM_EXIT_DEBUG;
+	kvm_run->debug.arch.pc = svm->vmcb->save.cs.base + svm->vmcb->save.rip;
+	kvm_run->debug.arch.exception = BP_VECTOR;
+	return 0;
+}
+
 static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run)
 {
 	int er;
@@ -2050,6 +2094,8 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm,
 	[SVM_EXIT_WRITE_DR3]			= emulate_on_interception,
 	[SVM_EXIT_WRITE_DR5]			= emulate_on_interception,
 	[SVM_EXIT_WRITE_DR7]			= emulate_on_interception,
+	[SVM_EXIT_EXCP_BASE + DB_VECTOR]	= db_interception,
+	[SVM_EXIT_EXCP_BASE + BP_VECTOR]	= bp_interception,
 	[SVM_EXIT_EXCP_BASE + UD_VECTOR]	= ud_interception,
 	[SVM_EXIT_EXCP_BASE + PF_VECTOR] 	= pf_interception,
 	[SVM_EXIT_EXCP_BASE + NM_VECTOR] 	= nm_interception,
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 1d974c1eaa7d..f55690ddb3ac 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -480,8 +480,13 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu)
 	eb = (1u << PF_VECTOR) | (1u << UD_VECTOR);
 	if (!vcpu->fpu_active)
 		eb |= 1u << NM_VECTOR;
-	if (vcpu->guest_debug.enabled)
-		eb |= 1u << DB_VECTOR;
+	if (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) {
+		if (vcpu->guest_debug &
+		    (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))
+			eb |= 1u << DB_VECTOR;
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP)
+			eb |= 1u << BP_VECTOR;
+	}
 	if (vcpu->arch.rmode.active)
 		eb = ~0;
 	if (vm_need_ept())
@@ -1003,40 +1008,23 @@ static void vmx_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg)
 	}
 }
 
-static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg)
+static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg)
 {
-	unsigned long dr7 = 0x400;
-	int old_singlestep;
-
-	old_singlestep = vcpu->guest_debug.singlestep;
-
-	vcpu->guest_debug.enabled = dbg->enabled;
-	if (vcpu->guest_debug.enabled) {
-		int i;
-
-		dr7 |= 0x200;  /* exact */
-		for (i = 0; i < 4; ++i) {
-			if (!dbg->breakpoints[i].enabled)
-				continue;
-			vcpu->guest_debug.bp[i] = dbg->breakpoints[i].address;
-			dr7 |= 2 << (i*2);    /* global enable */
-			dr7 |= 0 << (i*4+16); /* execution breakpoint */
-		}
-
-		vcpu->guest_debug.singlestep = dbg->singlestep;
-	} else
-		vcpu->guest_debug.singlestep = 0;
+	int old_debug = vcpu->guest_debug;
+	unsigned long flags;
 
-	if (old_singlestep && !vcpu->guest_debug.singlestep) {
-		unsigned long flags;
+	vcpu->guest_debug = dbg->control;
+	if (!(vcpu->guest_debug & KVM_GUESTDBG_ENABLE))
+		vcpu->guest_debug = 0;
 
-		flags = vmcs_readl(GUEST_RFLAGS);
+	flags = vmcs_readl(GUEST_RFLAGS);
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
+		flags |= X86_EFLAGS_TF | X86_EFLAGS_RF;
+	else if (old_debug & KVM_GUESTDBG_SINGLESTEP)
 		flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
-		vmcs_writel(GUEST_RFLAGS, flags);
-	}
+	vmcs_writel(GUEST_RFLAGS, flags);
 
 	update_exception_bitmap(vcpu);
-	vmcs_writel(GUEST_DR7, dr7);
 
 	return 0;
 }
@@ -2540,24 +2528,6 @@ static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr)
 	return 0;
 }
 
-static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu)
-{
-	struct kvm_guest_debug *dbg = &vcpu->guest_debug;
-
-	set_debugreg(dbg->bp[0], 0);
-	set_debugreg(dbg->bp[1], 1);
-	set_debugreg(dbg->bp[2], 2);
-	set_debugreg(dbg->bp[3], 3);
-
-	if (dbg->singlestep) {
-		unsigned long flags;
-
-		flags = vmcs_readl(GUEST_RFLAGS);
-		flags |= X86_EFLAGS_TF | X86_EFLAGS_RF;
-		vmcs_writel(GUEST_RFLAGS, flags);
-	}
-}
-
 static int handle_rmode_exception(struct kvm_vcpu *vcpu,
 				  int vec, u32 err_code)
 {
@@ -2574,9 +2544,17 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu,
 	 *        the required debugging infrastructure rework.
 	 */
 	switch (vec) {
-	case DE_VECTOR:
 	case DB_VECTOR:
+		if (vcpu->guest_debug &
+		    (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))
+			return 0;
+		kvm_queue_exception(vcpu, vec);
+		return 1;
 	case BP_VECTOR:
+		if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP)
+			return 0;
+		/* fall through */
+	case DE_VECTOR:
 	case OF_VECTOR:
 	case BR_VECTOR:
 	case UD_VECTOR:
@@ -2593,7 +2571,7 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu,
 static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 {
 	struct vcpu_vmx *vmx = to_vmx(vcpu);
-	u32 intr_info, error_code;
+	u32 intr_info, ex_no, error_code;
 	unsigned long cr2, rip;
 	u32 vect_info;
 	enum emulation_result er;
@@ -2653,14 +2631,16 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		return 1;
 	}
 
-	if ((intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK)) ==
-	    (INTR_TYPE_HARD_EXCEPTION | 1)) {
+	ex_no = intr_info & INTR_INFO_VECTOR_MASK;
+	if (ex_no == DB_VECTOR || ex_no == BP_VECTOR) {
 		kvm_run->exit_reason = KVM_EXIT_DEBUG;
-		return 0;
+		kvm_run->debug.arch.pc = vmcs_readl(GUEST_CS_BASE) + rip;
+		kvm_run->debug.arch.exception = ex_no;
+	} else {
+		kvm_run->exit_reason = KVM_EXIT_EXCEPTION;
+		kvm_run->ex.exception = ex_no;
+		kvm_run->ex.error_code = error_code;
 	}
-	kvm_run->exit_reason = KVM_EXIT_EXCEPTION;
-	kvm_run->ex.exception = intr_info & INTR_INFO_VECTOR_MASK;
-	kvm_run->ex.error_code = error_code;
 	return 0;
 }
 
@@ -3600,7 +3580,6 @@ static struct kvm_x86_ops vmx_x86_ops = {
 	.vcpu_put = vmx_vcpu_put,
 
 	.set_guest_debug = set_guest_debug,
-	.guest_debug_pre = kvm_guest_debug_pre,
 	.get_msr = vmx_get_msr,
 	.set_msr = vmx_set_msr,
 	.get_segment_base = vmx_get_segment_base,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index b5e9932e0f62..e990d164b56d 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3005,9 +3005,6 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 		goto out;
 	}
 
-	if (vcpu->guest_debug.enabled)
-		kvm_x86_ops->guest_debug_pre(vcpu);
-
 	vcpu->guest_mode = 1;
 	/*
 	 * Make sure that guest_mode assignment won't happen after
@@ -3218,7 +3215,7 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs)
 	/*
 	 * Don't leak debug flags in case they were set for guest debugging
 	 */
-	if (vcpu->guest_debug.enabled && vcpu->guest_debug.singlestep)
+	if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP)
 		regs->rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF);
 
 	vcpu_put(vcpu);
@@ -3837,8 +3834,8 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu,
 	return 0;
 }
 
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-				    struct kvm_debug_guest *dbg)
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg)
 {
 	int r;
 
@@ -3846,6 +3843,11 @@ int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
 
 	r = kvm_x86_ops->set_guest_debug(vcpu, dbg);
 
+	if (dbg->control & KVM_GUESTDBG_INJECT_DB)
+		kvm_queue_exception(vcpu, DB_VECTOR);
+	else if (dbg->control & KVM_GUESTDBG_INJECT_BP)
+		kvm_queue_exception(vcpu, BP_VECTOR);
+
 	vcpu_put(vcpu);
 
 	return r;
diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 0424326f1679..429a2ce202f9 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -126,6 +126,7 @@ struct kvm_run {
 			__u64 data_offset; /* relative to kvm_run start */
 		} io;
 		struct {
+			struct kvm_debug_exit_arch arch;
 		} debug;
 		/* KVM_EXIT_MMIO */
 		struct {
@@ -217,21 +218,6 @@ struct kvm_interrupt {
 	__u32 irq;
 };
 
-struct kvm_breakpoint {
-	__u32 enabled;
-	__u32 padding;
-	__u64 address;
-};
-
-/* for KVM_DEBUG_GUEST */
-struct kvm_debug_guest {
-	/* int */
-	__u32 enabled;
-	__u32 pad;
-	struct kvm_breakpoint breakpoints[4];
-	__u32 singlestep;
-};
-
 /* for KVM_GET_DIRTY_LOG */
 struct kvm_dirty_log {
 	__u32 slot;
@@ -292,6 +278,17 @@ struct kvm_s390_interrupt {
 	__u64 parm64;
 };
 
+/* for KVM_SET_GUEST_DEBUG */
+
+#define KVM_GUESTDBG_ENABLE		0x00000001
+#define KVM_GUESTDBG_SINGLESTEP		0x00000002
+
+struct kvm_guest_debug {
+	__u32 control;
+	__u32 pad;
+	struct kvm_guest_debug_arch arch;
+};
+
 #define KVM_TRC_SHIFT           16
 /*
  * kvm trace categories
@@ -396,6 +393,7 @@ struct kvm_trace_rec {
 #ifdef __KVM_HAVE_USER_NMI
 #define KVM_CAP_USER_NMI 22
 #endif
+#define KVM_CAP_SET_GUEST_DEBUG 23
 
 /*
  * ioctls for VM fds
@@ -440,7 +438,8 @@ struct kvm_trace_rec {
 #define KVM_SET_SREGS             _IOW(KVMIO,  0x84, struct kvm_sregs)
 #define KVM_TRANSLATE             _IOWR(KVMIO, 0x85, struct kvm_translation)
 #define KVM_INTERRUPT             _IOW(KVMIO,  0x86, struct kvm_interrupt)
-#define KVM_DEBUG_GUEST           _IOW(KVMIO,  0x87, struct kvm_debug_guest)
+/* KVM_DEBUG_GUEST is no longer supported, use KVM_SET_GUEST_DEBUG instead */
+#define KVM_DEBUG_GUEST           __KVM_DEPRECATED_DEBUG_GUEST
 #define KVM_GET_MSRS              _IOWR(KVMIO, 0x88, struct kvm_msrs)
 #define KVM_SET_MSRS              _IOW(KVMIO,  0x89, struct kvm_msrs)
 #define KVM_SET_CPUID             _IOW(KVMIO,  0x8a, struct kvm_cpuid)
@@ -469,6 +468,26 @@ struct kvm_trace_rec {
 #define KVM_SET_MP_STATE          _IOW(KVMIO,  0x99, struct kvm_mp_state)
 /* Available with KVM_CAP_NMI */
 #define KVM_NMI                   _IO(KVMIO,  0x9a)
+/* Available with KVM_CAP_SET_GUEST_DEBUG */
+#define KVM_SET_GUEST_DEBUG       _IOW(KVMIO,  0x9b, struct kvm_guest_debug)
+
+/*
+ * Deprecated interfaces
+ */
+struct kvm_breakpoint {
+	__u32 enabled;
+	__u32 padding;
+	__u64 address;
+};
+
+struct kvm_debug_guest {
+	__u32 enabled;
+	__u32 pad;
+	struct kvm_breakpoint breakpoints[4];
+	__u32 singlestep;
+};
+
+#define __KVM_DEPRECATED_DEBUG_GUEST _IOW(KVMIO,  0x87, struct kvm_debug_guest)
 
 #define KVM_TRC_INJ_VIRQ         (KVM_TRC_HANDLER + 0x02)
 #define KVM_TRC_REDELIVER_EVT    (KVM_TRC_HANDLER + 0x03)
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index bf6f703642fc..e92212f970db 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -73,7 +73,7 @@ struct kvm_vcpu {
 	struct kvm_run *run;
 	int guest_mode;
 	unsigned long requests;
-	struct kvm_guest_debug guest_debug;
+	unsigned long guest_debug;
 	int fpu_active;
 	int guest_fpu_loaded;
 	wait_queue_head_t wq;
@@ -255,8 +255,8 @@ int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
 				    struct kvm_mp_state *mp_state);
 int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
 				    struct kvm_mp_state *mp_state);
-int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu,
-				    struct kvm_debug_guest *dbg);
+int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
+					struct kvm_guest_debug *dbg);
 int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run);
 
 int kvm_arch_init(void *opaque);
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 29a667ce35b0..f83ef9c7e89b 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1755,13 +1755,13 @@ out_free2:
 		r = 0;
 		break;
 	}
-	case KVM_DEBUG_GUEST: {
-		struct kvm_debug_guest dbg;
+	case KVM_SET_GUEST_DEBUG: {
+		struct kvm_guest_debug dbg;
 
 		r = -EFAULT;
 		if (copy_from_user(&dbg, argp, sizeof dbg))
 			goto out;
-		r = kvm_arch_vcpu_ioctl_debug_guest(vcpu, &dbg);
+		r = kvm_arch_vcpu_ioctl_set_guest_debug(vcpu, &dbg);
 		if (r)
 			goto out;
 		r = 0;
-- 
cgit v1.2.3-55-g7522


From 989c0f0ed56468a4aa48711cef5acf122a40d1dd Mon Sep 17 00:00:00 2001
From: Jan Kiszka
Date: Thu, 18 Dec 2008 12:33:18 +0100
Subject: KVM: Remove old kvm_guest_debug structs

Remove the remaining arch fragments of the old guest debug interface
that now break non-x86 builds.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/ia64/include/asm/kvm_host.h    | 2 --
 arch/powerpc/include/asm/kvm_host.h | 6 ------
 arch/s390/include/asm/kvm_host.h    | 3 ---
 3 files changed, 11 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h
index 7da0c0963226..46f8b0eb5684 100644
--- a/arch/ia64/include/asm/kvm_host.h
+++ b/arch/ia64/include/asm/kvm_host.h
@@ -239,8 +239,6 @@ struct kvm_vm_data {
 
 struct kvm;
 struct kvm_vcpu;
-struct kvm_guest_debug{
-};
 
 struct kvm_mmio_req {
 	uint64_t addr;          /*  physical address		*/
diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h
index c1e436fe7738..a22600537a8c 100644
--- a/arch/powerpc/include/asm/kvm_host.h
+++ b/arch/powerpc/include/asm/kvm_host.h
@@ -200,10 +200,4 @@ struct kvm_vcpu_arch {
 	unsigned long pending_exceptions;
 };
 
-struct kvm_guest_debug {
-	int enabled;
-	unsigned long bp[4];
-	int singlestep;
-};
-
 #endif /* __POWERPC_KVM_HOST_H__ */
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 3c55e4107dcc..c6e674f5fca9 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -21,9 +21,6 @@
 /* memory slots that does not exposed to userspace */
 #define KVM_PRIVATE_MEM_SLOTS 4
 
-struct kvm_guest_debug {
-};
-
 struct sca_entry {
 	atomic_t scn;
 	__u64	reserved;
-- 
cgit v1.2.3-55-g7522


From 5d9b8e30f543a9f21a968a4cda71e8f6d1c66a61 Mon Sep 17 00:00:00 2001
From: Avi Kivity
Date: Sun, 4 Jan 2009 18:04:18 +0200
Subject: KVM: Add CONFIG_HAVE_KVM_IRQCHIP

Two KVM archs support irqchips and two don't.  Add a Kconfig item to
make selecting between the two models easier.

Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/ia64/kvm/Kconfig    | 4 ++++
 arch/powerpc/kvm/Kconfig | 3 +++
 arch/s390/kvm/Kconfig    | 3 +++
 arch/x86/kvm/Kconfig     | 4 ++++
 4 files changed, 14 insertions(+)

(limited to 'arch/s390')

diff --git a/arch/ia64/kvm/Kconfig b/arch/ia64/kvm/Kconfig
index f833a0b4188d..0a2d6b86075a 100644
--- a/arch/ia64/kvm/Kconfig
+++ b/arch/ia64/kvm/Kconfig
@@ -4,6 +4,10 @@
 config HAVE_KVM
 	bool
 
+config HAVE_KVM_IRQCHIP
+       bool
+       default y
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	depends on HAVE_KVM || IA64
diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig
index 146570550142..5a152a52796f 100644
--- a/arch/powerpc/kvm/Kconfig
+++ b/arch/powerpc/kvm/Kconfig
@@ -2,6 +2,9 @@
 # KVM configuration
 #
 
+config HAVE_KVM_IRQCHIP
+       bool
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	---help---
diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig
index e051cad1f1e0..3e260b7e37b2 100644
--- a/arch/s390/kvm/Kconfig
+++ b/arch/s390/kvm/Kconfig
@@ -4,6 +4,9 @@
 config HAVE_KVM
        bool
 
+config HAVE_KVM_IRQCHIP
+       bool
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	default y
diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig
index b81125f0bdee..0a303c3ed11f 100644
--- a/arch/x86/kvm/Kconfig
+++ b/arch/x86/kvm/Kconfig
@@ -4,6 +4,10 @@
 config HAVE_KVM
        bool
 
+config HAVE_KVM_IRQCHIP
+       bool
+       default y
+
 menuconfig VIRTUALIZATION
 	bool "Virtualization"
 	depends on HAVE_KVM || X86
-- 
cgit v1.2.3-55-g7522


From 2c411b48af3e3c534b9cfb6a79be1df384d1ca1a Mon Sep 17 00:00:00 2001
From: Christian Borntraeger
Date: Thu, 22 Jan 2009 14:20:27 +0100
Subject: KVM: s390: Fix printk on SIGP set arch

KVM on s390 does not support the ESA/390 architecture. We refuse to
change the architecture mode and print a warning. This patch removes
the printk for several reasons:

o A malicious guest can flood host dmesg
o The old message had no newline
o there is no connection between the message and the failing guest

This patch simply removes the printk. We already set the condition
code to 3 - the guest knows that something went wrong.

Reported-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/s390/kvm/sigp.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c
index 2a01b9e02801..f27dbedf0866 100644
--- a/arch/s390/kvm/sigp.c
+++ b/arch/s390/kvm/sigp.c
@@ -153,8 +153,6 @@ static int __sigp_set_arch(struct kvm_vcpu *vcpu, u32 parameter)
 
 	switch (parameter & 0xff) {
 	case 0:
-		printk(KERN_WARNING "kvm: request to switch to ESA/390 mode"
-							" not supported");
 		rc = 3; /* not operational */
 		break;
 	case 1:
-- 
cgit v1.2.3-55-g7522


From 70455a36a073cbb83ca17f92d135a6128c73cb3c Mon Sep 17 00:00:00 2001
From: Christian Borntraeger
Date: Thu, 22 Jan 2009 10:28:29 +0100
Subject: KVM: s390: Fix problem state check for b2 intercepts

The kernel handles some priviledged instruction exits. While I was
unable to trigger such an exit from guest userspace, the code should
check for supervisor state before emulating a priviledged instruction.

I also renamed kvm_s390_handle_priv to kvm_s390_handle_b2. After all
there are non priviledged b2 instructions like stck (store clock).

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/s390/kvm/intercept.c |  2 +-
 arch/s390/kvm/kvm-s390.h  |  2 +-
 arch/s390/kvm/priv.c      | 18 +++++++++++++++---
 3 files changed, 17 insertions(+), 5 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index 61236102203e..9d19803111ba 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -103,7 +103,7 @@ static int handle_lctl(struct kvm_vcpu *vcpu)
 static intercept_handler_t instruction_handlers[256] = {
 	[0x83] = kvm_s390_handle_diag,
 	[0xae] = kvm_s390_handle_sigp,
-	[0xb2] = kvm_s390_handle_priv,
+	[0xb2] = kvm_s390_handle_b2,
 	[0xb7] = handle_lctl,
 	[0xeb] = handle_lctlg,
 };
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index 3893cf12eacf..00bbe69b78da 100644
--- a/arch/s390/kvm/kvm-s390.h
+++ b/arch/s390/kvm/kvm-s390.h
@@ -50,7 +50,7 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code);
 
 /* implemented in priv.c */
-int kvm_s390_handle_priv(struct kvm_vcpu *vcpu);
+int kvm_s390_handle_b2(struct kvm_vcpu *vcpu);
 
 /* implemented in sigp.c */
 int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu);
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index 3605df45dd41..4b88834b8dd8 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -304,12 +304,24 @@ static intercept_handler_t priv_handlers[256] = {
 	[0xb1] = handle_stfl,
 };
 
-int kvm_s390_handle_priv(struct kvm_vcpu *vcpu)
+int kvm_s390_handle_b2(struct kvm_vcpu *vcpu)
 {
 	intercept_handler_t handler;
 
+	/*
+	 * a lot of B2 instructions are priviledged. We first check for
+	 * the priviledges ones, that we can handle in the kernel. If the
+	 * kernel can handle this instruction, we check for the problem
+	 * state bit and (a) handle the instruction or (b) send a code 2
+	 * program check.
+	 * Anything else goes to userspace.*/
 	handler = priv_handlers[vcpu->arch.sie_block->ipa & 0x00ff];
-	if (handler)
-		return handler(vcpu);
+	if (handler) {
+		if (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PSTATE)
+			return kvm_s390_inject_program_int(vcpu,
+						   PGM_PRIVILEGED_OPERATION);
+		else
+			return handler(vcpu);
+	}
 	return -ENOTSUPP;
 }
-- 
cgit v1.2.3-55-g7522


From b7e6e4d3602c738b8f61225d9f4514945df52f07 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger
Date: Thu, 22 Jan 2009 10:29:08 +0100
Subject: KVM: s390: Fix SIGP set prefix ioctl

This patch fixes the SET PREFIX interrupt if triggered by userspace.
Until now, it was not necessary, but life migration will need it. In
addition, it helped me creating SMP support for my kvm_crashme tool
(lets kvm execute random guest memory content).

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/s390/kvm/interrupt.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index f4fe28a2521a..0189356fe209 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -555,9 +555,14 @@ int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu,
 		VCPU_EVENT(vcpu, 3, "inject: program check %d (from user)",
 			   s390int->parm);
 		break;
+	case KVM_S390_SIGP_SET_PREFIX:
+		inti->prefix.address = s390int->parm;
+		inti->type = s390int->type;
+		VCPU_EVENT(vcpu, 3, "inject: set prefix to %x (from user)",
+			   s390int->parm);
+		break;
 	case KVM_S390_SIGP_STOP:
 	case KVM_S390_RESTART:
-	case KVM_S390_SIGP_SET_PREFIX:
 	case KVM_S390_INT_EMERGENCY:
 		VCPU_EVENT(vcpu, 3, "inject: type %x", s390int->type);
 		inti->type = s390int->type;
-- 
cgit v1.2.3-55-g7522


From d7fd5f1e3b195a8232b3ed768ac2809ddce8ca46 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger
Date: Thu, 26 Mar 2009 15:23:42 +0100
Subject: [S390] fix dump_stack vs. %p and (null)

The s390 implemenation of dump_stack uses %p to display stack content.
Since d97106ab53f812910a62d18afb9dbe882819c1ba (Make %p print '(null)'
for NULL pointers) this causes a strange output for dump_stack:

[...]
Process basename (pid: 8822, task: 00000000b2ece038, ksp: 00000000b24d7b38)
04000000b5685c00 00000000b24d7760 0000000000000002 (null)
00000000b24d7800 00000000b24d7778 00000000b24d7778 00000000001052fe
(null) 00000000b24d7b38 (null) 000000000000000a
000000000000000d (null) 00000000b24d7760 00000000b24d77d8
000000000051a7e8 00000000001052fe 00000000b24d7760 00000000b24d77b0
Call Trace:
[...]

This patch changes our dump_stack to use the appropriate %x format.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/traps.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/traps.c b/arch/s390/kernel/traps.c
index 4584d81984c0..c2e42cc65ce7 100644
--- a/arch/s390/kernel/traps.c
+++ b/arch/s390/kernel/traps.c
@@ -61,9 +61,11 @@ extern pgm_check_handler_t do_asce_exception;
 #define stack_pointer ({ void **sp; asm("la %0,0(15)" : "=&d" (sp)); sp; })
 
 #ifndef CONFIG_64BIT
+#define LONG "%08lx "
 #define FOURLONG "%08lx %08lx %08lx %08lx\n"
 static int kstack_depth_to_print = 12;
 #else /* CONFIG_64BIT */
+#define LONG "%016lx "
 #define FOURLONG "%016lx %016lx %016lx %016lx\n"
 static int kstack_depth_to_print = 20;
 #endif /* CONFIG_64BIT */
@@ -155,7 +157,7 @@ void show_stack(struct task_struct *task, unsigned long *sp)
 			break;
 		if (i && ((i * sizeof (long) % 32) == 0))
 			printk("\n       ");
-		printk("%p ", (void *)*stack++);
+		printk(LONG, *stack++);
 	}
 	printk("\n");
 	show_trace(task, sp);
-- 
cgit v1.2.3-55-g7522


From 099b765139929efdcf232f8870804accf8c4cdc5 Mon Sep 17 00:00:00 2001
From: Frank Munzert
Date: Thu, 26 Mar 2009 15:23:43 +0100
Subject: [S390] Automatic IPL after dump

Provide new shutdown action "dump_reipl" for automatic ipl after dump.

Signed-off-by: Frank Munzert <munzert@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
---
 arch/s390/include/asm/lowcore.h | 24 +++++------
 arch/s390/include/asm/system.h  | 16 ++++++++
 arch/s390/kernel/ipl.c          | 56 ++++++++++++++++++++++++--
 arch/s390/kernel/setup.c        |  5 +++
 arch/s390/kernel/smp.c          |  9 -----
 drivers/s390/char/zcore.c       | 88 ++++++++++++++++++++++++++++++++++++++++-
 6 files changed, 172 insertions(+), 26 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h
index f3720defdd16..ee4b10ff9387 100644
--- a/arch/s390/include/asm/lowcore.h
+++ b/arch/s390/include/asm/lowcore.h
@@ -111,7 +111,7 @@
 
 #define __LC_PASTE			0xE40
 
-#define __LC_PANIC_MAGIC		0xE00
+#define __LC_DUMP_REIPL			0xE00
 #ifndef __s390x__
 #define __LC_PFAULT_INTPARM             0x080
 #define __LC_CPU_TIMER_SAVE_AREA        0x0D8
@@ -286,12 +286,14 @@ struct _lowcore
 	__u64        int_clock;                /* 0xc98 */
         __u8         pad11[0xe00-0xca0];       /* 0xca0 */
 
-        /* 0xe00 is used as indicator for dump tools */
-        /* whether the kernel died with panic() or not */
-        __u32        panic_magic;              /* 0xe00 */
+	/* 0xe00 contains the address of the IPL Parameter */
+	/* Information block. Dump tools need IPIB for IPL */
+	/* after dump.					   */
+	__u32	     ipib;		       /* 0xe00 */
+	__u32	     ipib_checksum;	       /* 0xe04 */
 
         /* Align to the top 1k of prefix area */
-	__u8         pad12[0x1000-0xe04];      /* 0xe04 */
+	__u8	     pad12[0x1000-0xe08];      /* 0xe08 */
 #else /* !__s390x__ */
         /* prefix area: defined by architecture */
 	__u32        ccw1[2];                  /* 0x000 */
@@ -379,12 +381,14 @@ struct _lowcore
 	__u64        int_clock;                /* 0xde8 */
         __u8         pad12[0xe00-0xdf0];       /* 0xdf0 */
 
-        /* 0xe00 is used as indicator for dump tools */
-        /* whether the kernel died with panic() or not */
-        __u32        panic_magic;              /* 0xe00 */
+	/* 0xe00 contains the address of the IPL Parameter */
+	/* Information block. Dump tools need IPIB for IPL */
+	/* after dump.					   */
+	__u64	     ipib;		       /* 0xe00 */
+	__u32	     ipib_checksum;	       /* 0xe08 */
 
 	/* Per cpu primary space access list */
-	__u8	     pad_0xe04[0xe38-0xe04];   /* 0xe04 */
+	__u8	     pad_0xe0c[0xe38-0xe0c];   /* 0xe0c */
 	__u64	     vdso_per_cpu_data;	       /* 0xe38 */
 	__u32	     paste[16];		       /* 0xe40 */
 
@@ -433,8 +437,6 @@ static inline __u32 store_prefix(void)
 	return address;
 }
 
-#define __PANIC_MAGIC           0xDEADC0DE
-
 #endif
 
 #endif
diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h
index 3a8b26eb1f2e..3f2ccb82b863 100644
--- a/arch/s390/include/asm/system.h
+++ b/arch/s390/include/asm/system.h
@@ -458,6 +458,22 @@ static inline unsigned short stap(void)
 	return cpu_address;
 }
 
+static inline u32 cksm(void *addr, unsigned long len)
+{
+	register unsigned long _addr asm("0") = (unsigned long) addr;
+	register unsigned long _len asm("1") = len;
+	unsigned long accu = 0;
+
+	asm volatile(
+		"0:\n"
+		"	cksm	%0,%1\n"
+		"	jnz	0b\n"
+		: "+d" (accu), "+d" (_addr), "+d" (_len)
+		:
+		: "cc", "memory");
+	return accu;
+}
+
 extern void (*_machine_restart)(char *command);
 extern void (*_machine_halt)(void);
 extern void (*_machine_power_off)(void);
diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 2dcf590faba6..5663c1f8e46a 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -56,13 +56,14 @@ struct shutdown_trigger {
 };
 
 /*
- * Five shutdown action types are supported:
+ * The following shutdown action types are supported:
  */
 #define SHUTDOWN_ACTION_IPL_STR		"ipl"
 #define SHUTDOWN_ACTION_REIPL_STR	"reipl"
 #define SHUTDOWN_ACTION_DUMP_STR	"dump"
 #define SHUTDOWN_ACTION_VMCMD_STR	"vmcmd"
 #define SHUTDOWN_ACTION_STOP_STR	"stop"
+#define SHUTDOWN_ACTION_DUMP_REIPL_STR	"dump_reipl"
 
 struct shutdown_action {
 	char *name;
@@ -146,6 +147,7 @@ static enum ipl_method reipl_method = REIPL_METHOD_DEFAULT;
 static struct ipl_parameter_block *reipl_block_fcp;
 static struct ipl_parameter_block *reipl_block_ccw;
 static struct ipl_parameter_block *reipl_block_nss;
+static struct ipl_parameter_block *reipl_block_actual;
 
 static int dump_capabilities = DUMP_TYPE_NONE;
 static enum dump_type dump_type = DUMP_TYPE_NONE;
@@ -835,6 +837,7 @@ static int reipl_set_type(enum ipl_type type)
 			reipl_method = REIPL_METHOD_CCW_VM;
 		else
 			reipl_method = REIPL_METHOD_CCW_CIO;
+		reipl_block_actual = reipl_block_ccw;
 		break;
 	case IPL_TYPE_FCP:
 		if (diag308_set_works)
@@ -843,6 +846,7 @@ static int reipl_set_type(enum ipl_type type)
 			reipl_method = REIPL_METHOD_FCP_RO_VM;
 		else
 			reipl_method = REIPL_METHOD_FCP_RO_DIAG;
+		reipl_block_actual = reipl_block_fcp;
 		break;
 	case IPL_TYPE_FCP_DUMP:
 		reipl_method = REIPL_METHOD_FCP_DUMP;
@@ -852,6 +856,7 @@ static int reipl_set_type(enum ipl_type type)
 			reipl_method = REIPL_METHOD_NSS_DIAG;
 		else
 			reipl_method = REIPL_METHOD_NSS;
+		reipl_block_actual = reipl_block_nss;
 		break;
 	case IPL_TYPE_UNKNOWN:
 		reipl_method = REIPL_METHOD_DEFAULT;
@@ -1332,6 +1337,48 @@ static struct shutdown_action __refdata dump_action = {
 	.init	= dump_init,
 };
 
+static void dump_reipl_run(struct shutdown_trigger *trigger)
+{
+	preempt_disable();
+	/*
+	 * Bypass dynamic address translation (DAT) when storing IPL parameter
+	 * information block address and checksum into the prefix area
+	 * (corresponding to absolute addresses 0-8191).
+	 * When enhanced DAT applies and the STE format control in one,
+	 * the absolute address is formed without prefixing. In this case a
+	 * normal store (stg/st) into the prefix area would no more match to
+	 * absolute addresses 0-8191.
+	 */
+#ifdef CONFIG_64BIT
+	asm volatile("sturg %0,%1"
+		:: "a" ((unsigned long) reipl_block_actual),
+		"a" (&lowcore_ptr[smp_processor_id()]->ipib));
+#else
+	asm volatile("stura %0,%1"
+		:: "a" ((unsigned long) reipl_block_actual),
+		"a" (&lowcore_ptr[smp_processor_id()]->ipib));
+#endif
+	asm volatile("stura %0,%1"
+		:: "a" (cksm(reipl_block_actual, reipl_block_actual->hdr.len)),
+		"a" (&lowcore_ptr[smp_processor_id()]->ipib_checksum));
+	preempt_enable();
+	dump_run(trigger);
+}
+
+static int __init dump_reipl_init(void)
+{
+	if (!diag308_set_works)
+		return -EOPNOTSUPP;
+	else
+		return 0;
+}
+
+static struct shutdown_action __refdata dump_reipl_action = {
+	.name	= SHUTDOWN_ACTION_DUMP_REIPL_STR,
+	.fn	= dump_reipl_run,
+	.init	= dump_reipl_init,
+};
+
 /*
  * vmcmd shutdown action: Trigger vm command on shutdown.
  */
@@ -1421,7 +1468,8 @@ static struct shutdown_action stop_action = {SHUTDOWN_ACTION_STOP_STR,
 /* action list */
 
 static struct shutdown_action *shutdown_actions_list[] = {
-	&ipl_action, &reipl_action, &dump_action, &vmcmd_action, &stop_action};
+	&ipl_action, &reipl_action, &dump_reipl_action, &dump_action,
+	&vmcmd_action, &stop_action};
 #define SHUTDOWN_ACTIONS_COUNT (sizeof(shutdown_actions_list) / sizeof(void *))
 
 /*
@@ -1434,11 +1482,11 @@ static int set_trigger(const char *buf, struct shutdown_trigger *trigger,
 		       size_t len)
 {
 	int i;
+
 	for (i = 0; i < SHUTDOWN_ACTIONS_COUNT; i++) {
 		if (!shutdown_actions_list[i])
 			continue;
-		if (strncmp(buf, shutdown_actions_list[i]->name,
-			    strlen(shutdown_actions_list[i]->name)) == 0) {
+		if (sysfs_streq(buf, shutdown_actions_list[i]->name)) {
 			trigger->action = shutdown_actions_list[i];
 			return len;
 		}
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index c5cfb6185eac..8fdf08379ce9 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -86,6 +86,10 @@ volatile int __cpu_logical_map[NR_CPUS]; /* logical cpu to cpu address */
 int __initdata memory_end_set;
 unsigned long __initdata memory_end;
 
+/* An array with a pointer to the lowcore of every CPU. */
+struct _lowcore *lowcore_ptr[NR_CPUS];
+EXPORT_SYMBOL(lowcore_ptr);
+
 /*
  * This is set up by the setup-routine at boot-time
  * for S390 need to find out, what we have to setup
@@ -434,6 +438,7 @@ setup_lowcore(void)
 	lc->vdso_per_cpu_data = (unsigned long) &lc->paste[0];
 #endif
 	set_prefix((u32)(unsigned long) lc);
+	lowcore_ptr[0] = lc;
 }
 
 static void __init
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 2d337cbb9329..e279d0fbbbe8 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -50,12 +50,6 @@
 #include <asm/vdso.h>
 #include "entry.h"
 
-/*
- * An array with a pointer the lowcore of every CPU.
- */
-struct _lowcore *lowcore_ptr[NR_CPUS];
-EXPORT_SYMBOL(lowcore_ptr);
-
 static struct task_struct *current_set[NR_CPUS];
 
 static u8 smp_cpu_type;
@@ -82,9 +76,6 @@ void smp_send_stop(void)
 	/* Disable all interrupts/machine checks */
 	__load_psw_mask(psw_kernel_bits & ~PSW_MASK_MCHECK);
 
-	/* write magic number to zero page (absolute 0) */
-	lowcore_ptr[smp_processor_id()]->panic_magic = __PANIC_MAGIC;
-
 	/* stop all processors */
 	for_each_online_cpu(cpu) {
 		if (cpu == smp_processor_id())
diff --git a/drivers/s390/char/zcore.c b/drivers/s390/char/zcore.c
index eefc6611412e..cfe782ee6473 100644
--- a/drivers/s390/char/zcore.c
+++ b/drivers/s390/char/zcore.c
@@ -5,7 +5,7 @@
  *
  * For more information please refer to Documentation/s390/zfcpdump.txt
  *
- * Copyright IBM Corp. 2003,2007
+ * Copyright IBM Corp. 2003,2008
  * Author(s): Michael Holzheu
  */
 
@@ -48,12 +48,19 @@ struct sys_info {
 	union save_area	lc_mask;
 };
 
+struct ipib_info {
+	unsigned long	ipib;
+	u32		checksum;
+}  __attribute__((packed));
+
 static struct sys_info sys_info;
 static struct debug_info *zcore_dbf;
 static int hsa_available;
 static struct dentry *zcore_dir;
 static struct dentry *zcore_file;
 static struct dentry *zcore_memmap_file;
+static struct dentry *zcore_reipl_file;
+static struct ipl_parameter_block *ipl_block;
 
 /*
  * Copy memory from HSA to kernel or user memory (not reentrant):
@@ -527,6 +534,33 @@ static const struct file_operations zcore_memmap_fops = {
 	.release	= zcore_memmap_release,
 };
 
+static ssize_t zcore_reipl_write(struct file *filp, const char __user *buf,
+				 size_t count, loff_t *ppos)
+{
+	if (ipl_block) {
+		diag308(DIAG308_SET, ipl_block);
+		diag308(DIAG308_IPL, NULL);
+	}
+	return count;
+}
+
+static int zcore_reipl_open(struct inode *inode, struct file *filp)
+{
+	return 0;
+}
+
+static int zcore_reipl_release(struct inode *inode, struct file *filp)
+{
+	return 0;
+}
+
+static const struct file_operations zcore_reipl_fops = {
+	.owner		= THIS_MODULE,
+	.write		= zcore_reipl_write,
+	.open		= zcore_reipl_open,
+	.release	= zcore_reipl_release,
+};
+
 
 static void __init set_s390_lc_mask(union save_area *map)
 {
@@ -645,6 +679,39 @@ static int __init zcore_header_init(int arch, struct zcore_header *hdr)
 	return 0;
 }
 
+/*
+ * Provide IPL parameter information block from either HSA or memory
+ * for future reipl
+ */
+static int __init zcore_reipl_init(void)
+{
+	struct ipib_info ipib_info;
+	int rc;
+
+	rc = memcpy_hsa_kernel(&ipib_info, __LC_DUMP_REIPL, sizeof(ipib_info));
+	if (rc)
+		return rc;
+	if (ipib_info.ipib == 0)
+		return 0;
+	ipl_block = (void *) __get_free_page(GFP_KERNEL);
+	if (!ipl_block)
+		return -ENOMEM;
+	if (ipib_info.ipib < ZFCPDUMP_HSA_SIZE)
+		rc = memcpy_hsa_kernel(ipl_block, ipib_info.ipib, PAGE_SIZE);
+	else
+		rc = memcpy_real(ipl_block, ipib_info.ipib, PAGE_SIZE);
+	if (rc) {
+		free_page((unsigned long) ipl_block);
+		return rc;
+	}
+	if (cksm(ipl_block, ipl_block->hdr.len) != ipib_info.checksum) {
+		TRACE("Checksum does not match\n");
+		free_page((unsigned long) ipl_block);
+		ipl_block = NULL;
+	}
+	return 0;
+}
+
 static int __init zcore_init(void)
 {
 	unsigned char arch;
@@ -690,6 +757,10 @@ static int __init zcore_init(void)
 	if (rc)
 		goto fail;
 
+	rc = zcore_reipl_init();
+	if (rc)
+		goto fail;
+
 	zcore_dir = debugfs_create_dir("zcore" , NULL);
 	if (!zcore_dir) {
 		rc = -ENOMEM;
@@ -707,9 +778,17 @@ static int __init zcore_init(void)
 		rc = -ENOMEM;
 		goto fail_file;
 	}
+	zcore_reipl_file = debugfs_create_file("reipl", S_IRUSR, zcore_dir,
+						NULL, &zcore_reipl_fops);
+	if (!zcore_reipl_file) {
+		rc = -ENOMEM;
+		goto fail_memmap_file;
+	}
 	hsa_available = 1;
 	return 0;
 
+fail_memmap_file:
+	debugfs_remove(zcore_memmap_file);
 fail_file:
 	debugfs_remove(zcore_file);
 fail_dir:
@@ -723,10 +802,15 @@ static void __exit zcore_exit(void)
 {
 	debug_unregister(zcore_dbf);
 	sclp_sdias_exit();
+	free_page((unsigned long) ipl_block);
+	debugfs_remove(zcore_reipl_file);
+	debugfs_remove(zcore_memmap_file);
+	debugfs_remove(zcore_file);
+	debugfs_remove(zcore_dir);
 	diag308(DIAG308_REL_HSA, NULL);
 }
 
-MODULE_AUTHOR("Copyright IBM Corp. 2003,2007");
+MODULE_AUTHOR("Copyright IBM Corp. 2003,2008");
 MODULE_DESCRIPTION("zcore module for zfcpdump support");
 MODULE_LICENSE("GPL");
 
-- 
cgit v1.2.3-55-g7522


From 59fa4392dddae244a1148cbbcb090b5a5728f576 Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:23:44 +0100
Subject: [S390] page fault: invoke oom-killer

s390 arch backend for 1c0fe6e3bda0464728c23c8d84aa47567e8b716c
"mm: invoke oom-killer from page fault".

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/mm/fault.c | 28 ++--------------------------
 1 file changed, 2 insertions(+), 26 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index 4d537205e83c..833e8366c351 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -200,29 +200,6 @@ static void do_low_address(struct pt_regs *regs, unsigned long error_code)
 	do_no_context(regs, error_code, 0);
 }
 
-/*
- * We ran out of memory, or some other thing happened to us that made
- * us unable to handle the page fault gracefully.
- */
-static int do_out_of_memory(struct pt_regs *regs, unsigned long error_code,
-			    unsigned long address)
-{
-	struct task_struct *tsk = current;
-	struct mm_struct *mm = tsk->mm;
-
-	up_read(&mm->mmap_sem);
-	if (is_global_init(tsk)) {
-		yield();
-		down_read(&mm->mmap_sem);
-		return 1;
-	}
-	printk("VM: killing process %s\n", tsk->comm);
-	if (regs->psw.mask & PSW_MASK_PSTATE)
-		do_group_exit(SIGKILL);
-	do_no_context(regs, error_code, address);
-	return 0;
-}
-
 static void do_sigbus(struct pt_regs *regs, unsigned long error_code,
 		      unsigned long address)
 {
@@ -367,7 +344,6 @@ good_area:
 			goto bad_area;
 	}
 
-survive:
 	if (is_vm_hugetlb_page(vma))
 		address &= HPAGE_MASK;
 	/*
@@ -378,8 +354,8 @@ survive:
 	fault = handle_mm_fault(mm, vma, address, write);
 	if (unlikely(fault & VM_FAULT_ERROR)) {
 		if (fault & VM_FAULT_OOM) {
-			if (do_out_of_memory(regs, error_code, address))
-				goto survive;
+			up_read(&mm->mmap_sem);
+			pagefault_out_of_memory();
 			return;
 		} else if (fault & VM_FAULT_SIGBUS) {
 			do_sigbus(regs, error_code, address);
-- 
cgit v1.2.3-55-g7522


From b44b0ab3bac16356f03e94b1b49ba9305710c445 Mon Sep 17 00:00:00 2001
From: Stefan Weinhuber
Date: Thu, 26 Mar 2009 15:23:47 +0100
Subject: [S390] dasd: add large volume support

The dasd device driver will now support ECKD devices with more then
65520 cylinders.
In the traditional ECKD adressing scheme each track is addressed
by a 16-bit cylinder and 16-bit head number. The new addressing
scheme makes use of the fact that the actual number of heads is
never larger then 15, so 12 bits of the head number can be redefined
to be part of the cylinder address.

Signed-off-by: Stefan Weinhuber <wein@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/dasd.h    |  10 ++--
 arch/s390/include/asm/vtoc.h    |  16 ++++-
 drivers/s390/block/dasd_eckd.c  | 126 ++++++++++++++++++++++------------------
 drivers/s390/block/dasd_eckd.h  |   9 ++-
 drivers/s390/block/dasd_int.h   |   2 +-
 drivers/s390/block/dasd_ioctl.c |   4 +-
 drivers/s390/block/dasd_proc.c  |   2 +-
 fs/partitions/ibm.c             | 101 ++++++++++++++++++++++----------
 8 files changed, 171 insertions(+), 99 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/dasd.h b/arch/s390/include/asm/dasd.h
index e2db6f16d9c8..218bce81ec70 100644
--- a/arch/s390/include/asm/dasd.h
+++ b/arch/s390/include/asm/dasd.h
@@ -162,15 +162,15 @@ typedef struct dasd_profile_info_t {
         unsigned int dasd_io_nr_req[32]; /* histogram of # of requests in chanq */
 } dasd_profile_info_t;
 
-/* 
+/*
  * struct format_data_t
  * represents all data necessary to format a dasd
  */
 typedef struct format_data_t {
-	int start_unit; /* from track */
-	int stop_unit;  /* to track */
-	int blksize;    /* sectorsize */
-        int intensity;  
+	unsigned int start_unit; /* from track */
+	unsigned int stop_unit;  /* to track */
+	unsigned int blksize;	 /* sectorsize */
+	unsigned int intensity;
 } format_data_t;
 
 /*
diff --git a/arch/s390/include/asm/vtoc.h b/arch/s390/include/asm/vtoc.h
index 3a5267d90d29..8406a2b3157a 100644
--- a/arch/s390/include/asm/vtoc.h
+++ b/arch/s390/include/asm/vtoc.h
@@ -39,7 +39,7 @@ struct vtoc_labeldate
 	__u16 day;
 } __attribute__ ((packed));
 
-struct vtoc_volume_label
+struct vtoc_volume_label_cdl
 {
 	char volkey[4];		/* volume key = volume label */
 	char vollbl[4];		/* volume label */
@@ -56,6 +56,14 @@ struct vtoc_volume_label
 	char res3[29];		/* reserved */
 } __attribute__ ((packed));
 
+struct vtoc_volume_label_ldl {
+	char vollbl[4];		/* volume label */
+	char volid[6];		/* volume identifier */
+	char res3[69];		/* reserved */
+	char ldl_version;	/* version number, valid for ldl format */
+	__u64 formatted_blocks; /* valid when ldl_version >= f2  */
+} __attribute__ ((packed));
+
 struct vtoc_extent
 {
 	__u8 typeind;			/* extent type indicator */
@@ -140,7 +148,11 @@ struct vtoc_format4_label
 	char res2[10];		/* reserved */
 	__u8 DS4EFLVL;		/* extended free-space management level */
 	struct vtoc_cchhb DS4EFPTR; /* pointer to extended free-space info */
-	char res3[9];		/* reserved */
+	char res3;		/* reserved */
+	__u32 DS4DCYL;		/* number of logical cyls */
+	char res4[2];		/* reserved */
+	__u8 DS4DEVF2;		/* device flags */
+	char res5;		/* reserved */
 } __attribute__ ((packed));
 
 struct vtoc_ds5ext
diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c
index 0eb5e5888c42..69f93e626fd3 100644
--- a/drivers/s390/block/dasd_eckd.c
+++ b/drivers/s390/block/dasd_eckd.c
@@ -159,6 +159,14 @@ recs_per_track(struct dasd_eckd_characteristics * rdc,
 	return 0;
 }
 
+static void set_ch_t(struct ch_t *geo, __u32 cyl, __u8 head)
+{
+	geo->cyl = (__u16) cyl;
+	geo->head = cyl >> 16;
+	geo->head <<= 4;
+	geo->head |= head;
+}
+
 static int
 check_XRC (struct ccw1         *de_ccw,
            struct DE_eckd_data *data,
@@ -186,11 +194,12 @@ check_XRC (struct ccw1         *de_ccw,
 }
 
 static int
-define_extent(struct ccw1 * ccw, struct DE_eckd_data * data, int trk,
-	      int totrk, int cmd, struct dasd_device * device)
+define_extent(struct ccw1 *ccw, struct DE_eckd_data *data, unsigned int trk,
+	      unsigned int totrk, int cmd, struct dasd_device *device)
 {
 	struct dasd_eckd_private *private;
-	struct ch_t geo, beg, end;
+	u32 begcyl, endcyl;
+	u16 heads, beghead, endhead;
 	int rc = 0;
 
 	private = (struct dasd_eckd_private *) device->private;
@@ -248,27 +257,24 @@ define_extent(struct ccw1 * ccw, struct DE_eckd_data * data, int trk,
 	    && !(private->uses_cdl && trk < 2))
 		data->ga_extended |= 0x40; /* Regular Data Format Mode */
 
-	geo.cyl = private->rdc_data.no_cyl;
-	geo.head = private->rdc_data.trk_per_cyl;
-	beg.cyl = trk / geo.head;
-	beg.head = trk % geo.head;
-	end.cyl = totrk / geo.head;
-	end.head = totrk % geo.head;
+	heads = private->rdc_data.trk_per_cyl;
+	begcyl = trk / heads;
+	beghead = trk % heads;
+	endcyl = totrk / heads;
+	endhead = totrk % heads;
 
 	/* check for sequential prestage - enhance cylinder range */
 	if (data->attributes.operation == DASD_SEQ_PRESTAGE ||
 	    data->attributes.operation == DASD_SEQ_ACCESS) {
 
-		if (end.cyl + private->attrib.nr_cyl < geo.cyl)
-			end.cyl += private->attrib.nr_cyl;
+		if (endcyl + private->attrib.nr_cyl < private->real_cyl)
+			endcyl += private->attrib.nr_cyl;
 		else
-			end.cyl = (geo.cyl - 1);
+			endcyl = (private->real_cyl - 1);
 	}
 
-	data->beg_ext.cyl = beg.cyl;
-	data->beg_ext.head = beg.head;
-	data->end_ext.cyl = end.cyl;
-	data->end_ext.head = end.head;
+	set_ch_t(&data->beg_ext, begcyl, beghead);
+	set_ch_t(&data->end_ext, endcyl, endhead);
 	return rc;
 }
 
@@ -294,13 +300,14 @@ static int check_XRC_on_prefix(struct PFX_eckd_data *pfxdata,
 	return rc;
 }
 
-static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata, int trk,
-		  int totrk, int cmd, struct dasd_device *basedev,
-		  struct dasd_device *startdev)
+static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
+		  unsigned int trk, unsigned int totrk, int cmd,
+		  struct dasd_device *basedev, struct dasd_device *startdev)
 {
 	struct dasd_eckd_private *basepriv, *startpriv;
 	struct DE_eckd_data *data;
-	struct ch_t geo, beg, end;
+	u32 begcyl, endcyl;
+	u16 heads, beghead, endhead;
 	int rc = 0;
 
 	basepriv = (struct dasd_eckd_private *) basedev->private;
@@ -374,33 +381,30 @@ static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata, int trk,
 	    && !(basepriv->uses_cdl && trk < 2))
 		data->ga_extended |= 0x40; /* Regular Data Format Mode */
 
-	geo.cyl = basepriv->rdc_data.no_cyl;
-	geo.head = basepriv->rdc_data.trk_per_cyl;
-	beg.cyl = trk / geo.head;
-	beg.head = trk % geo.head;
-	end.cyl = totrk / geo.head;
-	end.head = totrk % geo.head;
+	heads = basepriv->rdc_data.trk_per_cyl;
+	begcyl = trk / heads;
+	beghead = trk % heads;
+	endcyl = totrk / heads;
+	endhead = totrk % heads;
 
 	/* check for sequential prestage - enhance cylinder range */
 	if (data->attributes.operation == DASD_SEQ_PRESTAGE ||
 	    data->attributes.operation == DASD_SEQ_ACCESS) {
 
-		if (end.cyl + basepriv->attrib.nr_cyl < geo.cyl)
-			end.cyl += basepriv->attrib.nr_cyl;
+		if (endcyl + basepriv->attrib.nr_cyl < basepriv->real_cyl)
+			endcyl += basepriv->attrib.nr_cyl;
 		else
-			end.cyl = (geo.cyl - 1);
+			endcyl = (basepriv->real_cyl - 1);
 	}
 
-	data->beg_ext.cyl = beg.cyl;
-	data->beg_ext.head = beg.head;
-	data->end_ext.cyl = end.cyl;
-	data->end_ext.head = end.head;
+	set_ch_t(&data->beg_ext, begcyl, beghead);
+	set_ch_t(&data->end_ext, endcyl, endhead);
 	return rc;
 }
 
 static void
-locate_record(struct ccw1 *ccw, struct LO_eckd_data *data, int trk,
-	      int rec_on_trk, int no_rec, int cmd,
+locate_record(struct ccw1 *ccw, struct LO_eckd_data *data, unsigned int trk,
+	      unsigned int rec_on_trk, int no_rec, int cmd,
 	      struct dasd_device * device, int reclen)
 {
 	struct dasd_eckd_private *private;
@@ -493,10 +497,11 @@ locate_record(struct ccw1 *ccw, struct LO_eckd_data *data, int trk,
 	default:
 		DEV_MESSAGE(KERN_ERR, device, "unknown opcode 0x%x", cmd);
 	}
-	data->seek_addr.cyl = data->search_arg.cyl =
-		trk / private->rdc_data.trk_per_cyl;
-	data->seek_addr.head = data->search_arg.head =
-		trk % private->rdc_data.trk_per_cyl;
+	set_ch_t(&data->seek_addr,
+		 trk / private->rdc_data.trk_per_cyl,
+		 trk % private->rdc_data.trk_per_cyl);
+	data->search_arg.cyl = data->seek_addr.cyl;
+	data->search_arg.head = data->seek_addr.head;
 	data->search_arg.record = rec_on_trk;
 }
 
@@ -1002,13 +1007,20 @@ dasd_eckd_check_characteristics(struct dasd_device *device)
 			    "rc=%d", rc);
 		goto out_err3;
 	}
+	/* find the vaild cylinder size */
+	if (private->rdc_data.no_cyl == LV_COMPAT_CYL &&
+	    private->rdc_data.long_no_cyl)
+		private->real_cyl = private->rdc_data.long_no_cyl;
+	else
+		private->real_cyl = private->rdc_data.no_cyl;
+
 	DEV_MESSAGE(KERN_INFO, device,
 		    "%04X/%02X(CU:%04X/%02X) Cyl:%d Head:%d Sec:%d",
 		    private->rdc_data.dev_type,
 		    private->rdc_data.dev_model,
 		    private->rdc_data.cu_type,
 		    private->rdc_data.cu_model.model,
-		    private->rdc_data.no_cyl,
+		    private->real_cyl,
 		    private->rdc_data.trk_per_cyl,
 		    private->rdc_data.sec_per_trk);
 	return 0;
@@ -1157,8 +1169,6 @@ dasd_eckd_end_analysis(struct dasd_block *block)
 	}
 
 	private->uses_cdl = 1;
-	/* Calculate number of blocks/records per track. */
-	blk_per_trk = recs_per_track(&private->rdc_data, 0, block->bp_block);
 	/* Check Track 0 for Compatible Disk Layout */
 	count_area = NULL;
 	for (i = 0; i < 3; i++) {
@@ -1200,14 +1210,14 @@ dasd_eckd_end_analysis(struct dasd_block *block)
 		block->s2b_shift++;
 
 	blk_per_trk = recs_per_track(&private->rdc_data, 0, block->bp_block);
-	block->blocks = (private->rdc_data.no_cyl *
+	block->blocks = (private->real_cyl *
 			  private->rdc_data.trk_per_cyl *
 			  blk_per_trk);
 
 	DEV_MESSAGE(KERN_INFO, device,
 		    "(%dkB blks): %dkB at %dkB/trk %s",
 		    (block->bp_block >> 10),
-		    ((private->rdc_data.no_cyl *
+		    ((private->real_cyl *
 		      private->rdc_data.trk_per_cyl *
 		      blk_per_trk * (block->bp_block >> 9)) >> 1),
 		    ((blk_per_trk * block->bp_block) >> 10),
@@ -1262,7 +1272,8 @@ dasd_eckd_format_device(struct dasd_device * device,
 	struct eckd_count *ect;
 	struct ccw1 *ccw;
 	void *data;
-	int rpt, cyl, head;
+	int rpt;
+	struct ch_t address;
 	int cplength, datasize;
 	int i;
 	int intensity = 0;
@@ -1270,24 +1281,25 @@ dasd_eckd_format_device(struct dasd_device * device,
 
 	private = (struct dasd_eckd_private *) device->private;
 	rpt = recs_per_track(&private->rdc_data, 0, fdata->blksize);
-	cyl = fdata->start_unit / private->rdc_data.trk_per_cyl;
-	head = fdata->start_unit % private->rdc_data.trk_per_cyl;
+	set_ch_t(&address,
+		 fdata->start_unit / private->rdc_data.trk_per_cyl,
+		 fdata->start_unit % private->rdc_data.trk_per_cyl);
 
 	/* Sanity checks. */
 	if (fdata->start_unit >=
-	    (private->rdc_data.no_cyl * private->rdc_data.trk_per_cyl)) {
-		DEV_MESSAGE(KERN_INFO, device, "Track no %d too big!",
+	    (private->real_cyl * private->rdc_data.trk_per_cyl)) {
+		DEV_MESSAGE(KERN_INFO, device, "Track no %u too big!",
 			    fdata->start_unit);
 		return ERR_PTR(-EINVAL);
 	}
 	if (fdata->start_unit > fdata->stop_unit) {
-		DEV_MESSAGE(KERN_INFO, device, "Track %d reached! ending.",
+		DEV_MESSAGE(KERN_INFO, device, "Track %u reached! ending.",
 			    fdata->start_unit);
 		return ERR_PTR(-EINVAL);
 	}
 	if (dasd_check_blocksize(fdata->blksize) != 0) {
 		DEV_MESSAGE(KERN_WARNING, device,
-			    "Invalid blocksize %d...terminating!",
+			    "Invalid blocksize %u...terminating!",
 			    fdata->blksize);
 		return ERR_PTR(-EINVAL);
 	}
@@ -1389,8 +1401,8 @@ dasd_eckd_format_device(struct dasd_device * device,
 	if (intensity & 0x01) {	/* write record zero */
 		ect = (struct eckd_count *) data;
 		data += sizeof(struct eckd_count);
-		ect->cyl = cyl;
-		ect->head = head;
+		ect->cyl = address.cyl;
+		ect->head = address.head;
 		ect->record = 0;
 		ect->kl = 0;
 		ect->dl = 8;
@@ -1404,8 +1416,8 @@ dasd_eckd_format_device(struct dasd_device * device,
 	if ((intensity & ~0x08) & 0x04) {	/* erase track */
 		ect = (struct eckd_count *) data;
 		data += sizeof(struct eckd_count);
-		ect->cyl = cyl;
-		ect->head = head;
+		ect->cyl = address.cyl;
+		ect->head = address.head;
 		ect->record = 1;
 		ect->kl = 0;
 		ect->dl = 0;
@@ -1418,8 +1430,8 @@ dasd_eckd_format_device(struct dasd_device * device,
 		for (i = 0; i < rpt; i++) {
 			ect = (struct eckd_count *) data;
 			data += sizeof(struct eckd_count);
-			ect->cyl = cyl;
-			ect->head = head;
+			ect->cyl = address.cyl;
+			ect->head = address.head;
 			ect->record = i + 1;
 			ect->kl = 0;
 			ect->dl = fdata->blksize;
diff --git a/drivers/s390/block/dasd_eckd.h b/drivers/s390/block/dasd_eckd.h
index 2476f87d21d0..eecfa776db15 100644
--- a/drivers/s390/block/dasd_eckd.h
+++ b/drivers/s390/block/dasd_eckd.h
@@ -48,6 +48,11 @@
 #define PSF_ORDER_PRSSD 0x18
 #define PSF_ORDER_SSC	0x1D
 
+/*
+ * Size that is reportet for large volumes in the old 16-bit no_cyl field
+ */
+#define LV_COMPAT_CYL 0xFFFE
+
 /*****************************************************************************
  * SECTION: Type Definitions
  ****************************************************************************/
@@ -228,7 +233,8 @@ struct dasd_eckd_characteristics {
 	__u8 factor7;
 	__u8 factor8;
 	__u8 reserved2[3];
-	__u8 reserved3[10];
+	__u8 reserved3[6];
+	__u32 long_no_cyl;
 } __attribute__ ((packed));
 
 /* elements of the configuration data */
@@ -406,6 +412,7 @@ struct dasd_eckd_private {
 	int uses_cdl;
 	struct attrib_data_t attrib;	/* e.g. cache operations */
 	struct dasd_rssd_features features;
+	u32 real_cyl;
 
 	/* alias managemnet */
 	struct dasd_uid uid;
diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h
index 4a39084d9c95..29991a9fa07a 100644
--- a/drivers/s390/block/dasd_int.h
+++ b/drivers/s390/block/dasd_int.h
@@ -378,7 +378,7 @@ struct dasd_block {
 	struct block_device *bdev;
 	atomic_t open_count;
 
-	unsigned long blocks;	   /* size of volume in blocks */
+	unsigned long long blocks; /* size of volume in blocks */
 	unsigned int bp_block;	   /* bytes per block */
 	unsigned int s2b_shift;	   /* log2 (bp_block/512) */
 
diff --git a/drivers/s390/block/dasd_ioctl.c b/drivers/s390/block/dasd_ioctl.c
index 16e6ba462cb6..a3bbdb807bad 100644
--- a/drivers/s390/block/dasd_ioctl.c
+++ b/drivers/s390/block/dasd_ioctl.c
@@ -146,7 +146,7 @@ static int dasd_format(struct dasd_block *block, struct format_data_t *fdata)
 	}
 
 	DBF_DEV_EVENT(DBF_NOTICE, base,
-		      "formatting units %d to %d (%d B blocks) flags %d",
+		      "formatting units %u to %u (%u B blocks) flags %u",
 		      fdata->start_unit,
 		      fdata->stop_unit, fdata->blksize, fdata->intensity);
 
@@ -170,7 +170,7 @@ static int dasd_format(struct dasd_block *block, struct format_data_t *fdata)
 		if (rc) {
 			if (rc != -ERESTARTSYS)
 				DEV_MESSAGE(KERN_ERR, base,
-					    " Formatting of unit %d failed "
+					    " Formatting of unit %u failed "
 					    "with rc = %d",
 					    fdata->start_unit, rc);
 			return rc;
diff --git a/drivers/s390/block/dasd_proc.c b/drivers/s390/block/dasd_proc.c
index bf6fd348f20e..0aa569419d57 100644
--- a/drivers/s390/block/dasd_proc.c
+++ b/drivers/s390/block/dasd_proc.c
@@ -112,7 +112,7 @@ dasd_devices_show(struct seq_file *m, void *v)
 			seq_printf(m, "n/f	 ");
 		else
 			seq_printf(m,
-				   "at blocksize: %d, %ld blocks, %ld MB",
+				   "at blocksize: %d, %lld blocks, %lld MB",
 				   block->bp_block, block->blocks,
 				   ((block->bp_block >> 9) *
 				    block->blocks) >> 11);
diff --git a/fs/partitions/ibm.c b/fs/partitions/ibm.c
index 1e064c4a4f86..46297683cd34 100644
--- a/fs/partitions/ibm.c
+++ b/fs/partitions/ibm.c
@@ -21,20 +21,38 @@
  * compute the block number from a
  * cyl-cyl-head-head structure
  */
-static inline int
+static sector_t
 cchh2blk (struct vtoc_cchh *ptr, struct hd_geometry *geo) {
-        return ptr->cc * geo->heads * geo->sectors +
-	       ptr->hh * geo->sectors;
+
+	sector_t cyl;
+	__u16 head;
+
+	/*decode cylinder and heads for large volumes */
+	cyl = ptr->hh & 0xFFF0;
+	cyl <<= 12;
+	cyl |= ptr->cc;
+	head = ptr->hh & 0x000F;
+	return cyl * geo->heads * geo->sectors +
+	       head * geo->sectors;
 }
 
 /*
  * compute the block number from a
  * cyl-cyl-head-head-block structure
  */
-static inline int
+static sector_t
 cchhb2blk (struct vtoc_cchhb *ptr, struct hd_geometry *geo) {
-        return ptr->cc * geo->heads * geo->sectors +
-		ptr->hh * geo->sectors +
+
+	sector_t cyl;
+	__u16 head;
+
+	/*decode cylinder and heads for large volumes */
+	cyl = ptr->hh & 0xFFF0;
+	cyl <<= 12;
+	cyl |= ptr->cc;
+	head = ptr->hh & 0x000F;
+	return	cyl * geo->heads * geo->sectors +
+		head * geo->sectors +
 		ptr->b;
 }
 
@@ -43,14 +61,15 @@ cchhb2blk (struct vtoc_cchhb *ptr, struct hd_geometry *geo) {
 int
 ibm_partition(struct parsed_partitions *state, struct block_device *bdev)
 {
-	int blocksize, offset, size,res;
-	loff_t i_size;
+	int blocksize, res;
+	loff_t i_size, offset, size, fmt_size;
 	dasd_information2_t *info;
 	struct hd_geometry *geo;
 	char type[5] = {0,};
 	char name[7] = {0,};
 	union label_t {
-		struct vtoc_volume_label vol;
+		struct vtoc_volume_label_cdl vol;
+		struct vtoc_volume_label_ldl lnx;
 		struct vtoc_cms_label cms;
 	} *label;
 	unsigned char *data;
@@ -85,14 +104,16 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev)
 	if (data == NULL)
 		goto out_readerr;
 
-	strncpy (type, data, 4);
-	if ((!info->FBA_layout) && (!strcmp(info->type, "ECKD")))
-		strncpy(name, data + 8, 6);
-	else
-		strncpy(name, data + 4, 6);
 	memcpy(label, data, sizeof(union label_t));
 	put_dev_sector(sect);
 
+	if ((!info->FBA_layout) && (!strcmp(info->type, "ECKD"))) {
+		strncpy(type, label->vol.vollbl, 4);
+		strncpy(name, label->vol.volid, 6);
+	} else {
+		strncpy(type, label->lnx.vollbl, 4);
+		strncpy(name, label->lnx.volid, 6);
+	}
 	EBCASC(type, 4);
 	EBCASC(name, 6);
 
@@ -110,36 +131,54 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev)
 			/*
 			 * VM style CMS1 labeled disk
 			 */
+			blocksize = label->cms.block_size;
 			if (label->cms.disk_offset != 0) {
 				printk("CMS1/%8s(MDSK):", name);
 				/* disk is reserved minidisk */
-				blocksize = label->cms.block_size;
 				offset = label->cms.disk_offset;
 				size = (label->cms.block_count - 1)
 					* (blocksize >> 9);
 			} else {
 				printk("CMS1/%8s:", name);
 				offset = (info->label_block + 1);
-				size = i_size >> 9;
+				size = label->cms.block_count
+					* (blocksize >> 9);
 			}
+			put_partition(state, 1, offset*(blocksize >> 9),
+				      size-offset*(blocksize >> 9));
 		} else {
-			/*
-			 * Old style LNX1 or unlabeled disk
-			 */
-			if (strncmp(type, "LNX1", 4) == 0)
-				printk ("LNX1/%8s:", name);
-			else
+			if (strncmp(type, "LNX1", 4) == 0) {
+				printk("LNX1/%8s:", name);
+				if (label->lnx.ldl_version == 0xf2) {
+					fmt_size = label->lnx.formatted_blocks
+						* (blocksize >> 9);
+				} else if (!strcmp(info->type, "ECKD")) {
+					/* formated w/o large volume support */
+					fmt_size = geo->cylinders * geo->heads
+					      * geo->sectors * (blocksize >> 9);
+				} else {
+					/* old label and no usable disk geometry
+					 * (e.g. DIAG) */
+					fmt_size = i_size >> 9;
+				}
+				size = i_size >> 9;
+				if (fmt_size < size)
+					size = fmt_size;
+				offset = (info->label_block + 1);
+			} else {
+				/* unlabeled disk */
 				printk("(nonl)");
-			offset = (info->label_block + 1);
-			size = i_size >> 9;
-		}
-		put_partition(state, 1, offset*(blocksize >> 9),
+				size = i_size >> 9;
+				offset = (info->label_block + 1);
+			}
+			put_partition(state, 1, offset*(blocksize >> 9),
 				      size-offset*(blocksize >> 9));
+		}
 	} else if (info->format == DASD_FORMAT_CDL) {
 		/*
 		 * New style CDL formatted disk
 		 */
-		unsigned int blk;
+		sector_t blk;
 		int counter;
 
 		/*
@@ -166,7 +205,8 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev)
 				/* skip FMT4 / FMT5 / FMT7 labels */
 				if (f1.DS1FMTID == _ascebc['4']
 				    || f1.DS1FMTID == _ascebc['5']
-				    || f1.DS1FMTID == _ascebc['7']) {
+				    || f1.DS1FMTID == _ascebc['7']
+				    || f1.DS1FMTID == _ascebc['9']) {
 					blk++;
 					data = read_dev_sector(bdev, blk *
 							       (blocksize/512),
@@ -174,8 +214,9 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev)
 					continue;
 				}
 
-				/* only FMT1 valid at this point */
-				if (f1.DS1FMTID != _ascebc['1'])
+				/* only FMT1 and 8 labels valid at this point */
+				if (f1.DS1FMTID != _ascebc['1'] &&
+				    f1.DS1FMTID != _ascebc['8'])
 					break;
 
 				/* OK, we got valid partition data */
-- 
cgit v1.2.3-55-g7522


From f3eb5384cf0325c02e306b1d81e70f81a03d7432 Mon Sep 17 00:00:00 2001
From: Stefan Weinhuber
Date: Thu, 26 Mar 2009 15:23:48 +0100
Subject: [S390] dasd: add High Performance FICON support

To support High Performance FICON, the DASD device driver has to
translate I/O requests into the new transport mode control words (TCW)
instead of the traditional (command mode) CCW requests.

Signed-off-by: Stefan Weinhuber <wein@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/idals.h      |  17 +-
 drivers/s390/block/dasd.c          |  70 ++-
 drivers/s390/block/dasd_3990_erp.c | 141 ++++--
 drivers/s390/block/dasd_alias.c    |  25 +-
 drivers/s390/block/dasd_devmap.c   |   7 +
 drivers/s390/block/dasd_eckd.c     | 972 +++++++++++++++++++++++++++++++++----
 drivers/s390/block/dasd_eckd.h     |  40 +-
 drivers/s390/block/dasd_eer.c      |  21 +-
 drivers/s390/block/dasd_int.h      |   5 +-
 9 files changed, 1123 insertions(+), 175 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/idals.h b/arch/s390/include/asm/idals.h
index e82c10efe65a..aae276d00383 100644
--- a/arch/s390/include/asm/idals.h
+++ b/arch/s390/include/asm/idals.h
@@ -44,24 +44,18 @@ idal_is_needed(void *vaddr, unsigned int length)
 /*
  * Return the number of idal words needed for an address/length pair.
  */
-static inline unsigned int
-idal_nr_words(void *vaddr, unsigned int length)
+static inline unsigned int idal_nr_words(void *vaddr, unsigned int length)
 {
-#ifdef __s390x__
-	if (idal_is_needed(vaddr, length))
-		return ((__pa(vaddr) & (IDA_BLOCK_SIZE-1)) + length + 
-			(IDA_BLOCK_SIZE-1)) >> IDA_SIZE_LOG;
-#endif
-	return 0;
+	return ((__pa(vaddr) & (IDA_BLOCK_SIZE-1)) + length +
+		(IDA_BLOCK_SIZE-1)) >> IDA_SIZE_LOG;
 }
 
 /*
  * Create the list of idal words for an address/length pair.
  */
-static inline unsigned long *
-idal_create_words(unsigned long *idaws, void *vaddr, unsigned int length)
+static inline unsigned long *idal_create_words(unsigned long *idaws,
+					       void *vaddr, unsigned int length)
 {
-#ifdef __s390x__
 	unsigned long paddr;
 	unsigned int cidaw;
 
@@ -74,7 +68,6 @@ idal_create_words(unsigned long *idaws, void *vaddr, unsigned int length)
 		paddr += IDA_BLOCK_SIZE;
 		*idaws++ = paddr;
 	}
-#endif
 	return idaws;
 }
 
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index 93972ed7f2df..00f7d24b337a 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -22,6 +22,7 @@
 #include <asm/ebcdic.h>
 #include <asm/idals.h>
 #include <asm/todclk.h>
+#include <asm/itcw.h>
 
 /* This is ugly... */
 #define PRINTK_HEADER "dasd:"
@@ -852,8 +853,13 @@ int dasd_start_IO(struct dasd_ccw_req *cqr)
 	cqr->startclk = get_clock();
 	cqr->starttime = jiffies;
 	cqr->retries--;
-	rc = ccw_device_start(device->cdev, cqr->cpaddr, (long) cqr,
-			      cqr->lpm, 0);
+	if (cqr->cpmode == 1) {
+		rc = ccw_device_tm_start(device->cdev, cqr->cpaddr,
+					 (long) cqr, cqr->lpm);
+	} else {
+		rc = ccw_device_start(device->cdev, cqr->cpaddr,
+				      (long) cqr, cqr->lpm, 0);
+	}
 	switch (rc) {
 	case 0:
 		cqr->status = DASD_CQR_IN_IO;
@@ -881,9 +887,12 @@ int dasd_start_IO(struct dasd_ccw_req *cqr)
 			      " retry on all pathes");
 		break;
 	case -ENODEV:
+		DBF_DEV_EVENT(DBF_DEBUG, device, "%s",
+			      "start_IO: -ENODEV device gone, retry");
+		break;
 	case -EIO:
 		DBF_DEV_EVENT(DBF_ERR, device, "%s",
-			      "start_IO: device gone, retry");
+			      "start_IO: -EIO device gone, retry");
 		break;
 	default:
 		DEV_MESSAGE(KERN_ERR, device,
@@ -1015,9 +1024,9 @@ void dasd_int_handler(struct ccw_device *cdev, unsigned long intparm,
 
 	/* check for unsolicited interrupts */
 	cqr = (struct dasd_ccw_req *) intparm;
-	if (!cqr || ((irb->scsw.cmd.cc == 1) &&
-		     (irb->scsw.cmd.fctl & SCSW_FCTL_START_FUNC) &&
-		     (irb->scsw.cmd.stctl & SCSW_STCTL_STATUS_PEND))) {
+	if (!cqr || ((scsw_cc(&irb->scsw) == 1) &&
+		     (scsw_fctl(&irb->scsw) & SCSW_FCTL_START_FUNC) &&
+		     (scsw_stctl(&irb->scsw) & SCSW_STCTL_STATUS_PEND))) {
 		if (cqr && cqr->status == DASD_CQR_IN_IO)
 			cqr->status = DASD_CQR_QUEUED;
 		device = dasd_device_from_cdev_locked(cdev);
@@ -1040,7 +1049,7 @@ void dasd_int_handler(struct ccw_device *cdev, unsigned long intparm,
 
 	/* Check for clear pending */
 	if (cqr->status == DASD_CQR_CLEAR_PENDING &&
-	    irb->scsw.cmd.fctl & SCSW_FCTL_CLEAR_FUNC) {
+	    scsw_fctl(&irb->scsw) & SCSW_FCTL_CLEAR_FUNC) {
 		cqr->status = DASD_CQR_CLEARED;
 		dasd_device_clear_timer(device);
 		wake_up(&dasd_flush_wq);
@@ -1048,7 +1057,7 @@ void dasd_int_handler(struct ccw_device *cdev, unsigned long intparm,
 		return;
 	}
 
- 	/* check status - the request might have been killed by dyn detach */
+	/* check status - the request might have been killed by dyn detach */
 	if (cqr->status != DASD_CQR_IN_IO) {
 		MESSAGE(KERN_DEBUG,
 			"invalid status: bus_id %s, status %02x",
@@ -1059,8 +1068,8 @@ void dasd_int_handler(struct ccw_device *cdev, unsigned long intparm,
 		      ((irb->scsw.cmd.cstat << 8) | irb->scsw.cmd.dstat), cqr);
 	next = NULL;
 	expires = 0;
-	if (irb->scsw.cmd.dstat == (DEV_STAT_CHN_END | DEV_STAT_DEV_END) &&
-	    irb->scsw.cmd.cstat == 0 && !irb->esw.esw0.erw.cons) {
+	if (scsw_dstat(&irb->scsw) == (DEV_STAT_CHN_END | DEV_STAT_DEV_END) &&
+	    scsw_cstat(&irb->scsw) == 0) {
 		/* request was completed successfully */
 		cqr->status = DASD_CQR_SUCCESS;
 		cqr->stopclk = now;
@@ -1991,8 +2000,11 @@ static void dasd_setup_queue(struct dasd_block *block)
 	blk_queue_max_sectors(block->request_queue, max);
 	blk_queue_max_phys_segments(block->request_queue, -1L);
 	blk_queue_max_hw_segments(block->request_queue, -1L);
-	blk_queue_max_segment_size(block->request_queue, -1L);
-	blk_queue_segment_boundary(block->request_queue, -1L);
+	/* with page sized segments we can translate each segement into
+	 * one idaw/tidaw
+	 */
+	blk_queue_max_segment_size(block->request_queue, PAGE_SIZE);
+	blk_queue_segment_boundary(block->request_queue, PAGE_SIZE - 1);
 	blk_queue_ordered(block->request_queue, QUEUE_ORDERED_DRAIN, NULL);
 }
 
@@ -2432,6 +2444,40 @@ int dasd_generic_read_dev_chars(struct dasd_device *device, char *magic,
 }
 EXPORT_SYMBOL_GPL(dasd_generic_read_dev_chars);
 
+/*
+ *   In command mode and transport mode we need to look for sense
+ *   data in different places. The sense data itself is allways
+ *   an array of 32 bytes, so we can unify the sense data access
+ *   for both modes.
+ */
+char *dasd_get_sense(struct irb *irb)
+{
+	struct tsb *tsb = NULL;
+	char *sense = NULL;
+
+	if (scsw_is_tm(&irb->scsw) && (irb->scsw.tm.fcxs == 0x01)) {
+		if (irb->scsw.tm.tcw)
+			tsb = tcw_get_tsb((struct tcw *)(unsigned long)
+					  irb->scsw.tm.tcw);
+		if (tsb && tsb->length == 64 && tsb->flags)
+			switch (tsb->flags & 0x07) {
+			case 1:	/* tsa_iostat */
+				sense = tsb->tsa.iostat.sense;
+				break;
+			case 2: /* tsa_ddpc */
+				sense = tsb->tsa.ddpc.sense;
+				break;
+			default:
+				/* currently we don't use interrogate data */
+				break;
+			}
+	} else if (irb->esw.esw0.erw.cons) {
+		sense = irb->ecw;
+	}
+	return sense;
+}
+EXPORT_SYMBOL_GPL(dasd_get_sense);
+
 static int __init dasd_init(void)
 {
 	int rc;
diff --git a/drivers/s390/block/dasd_3990_erp.c b/drivers/s390/block/dasd_3990_erp.c
index d82aad5224f0..4cee45916144 100644
--- a/drivers/s390/block/dasd_3990_erp.c
+++ b/drivers/s390/block/dasd_3990_erp.c
@@ -1561,6 +1561,13 @@ dasd_3990_erp_action_1B_32(struct dasd_ccw_req * default_erp, char *sense)
 		cqr = cqr->refers;
 	}
 
+	if (scsw_is_tm(&cqr->irb.scsw)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "32 bit sense, action 1B is not defined"
+			      " in transport mode - just retry");
+		return default_erp;
+	}
+
 	/* for imprecise ending just do default erp */
 	if (sense[1] & 0x01) {
 
@@ -1599,7 +1606,7 @@ dasd_3990_erp_action_1B_32(struct dasd_ccw_req * default_erp, char *sense)
 	oldccw = cqr->cpaddr;
 	if (oldccw->cmd_code == DASD_ECKD_CCW_PFX) {
 		PFX_data = cqr->data;
-		memcpy(DE_data, &PFX_data->define_extend,
+		memcpy(DE_data, &PFX_data->define_extent,
 		       sizeof(struct DE_eckd_data));
 	} else
 		memcpy(DE_data, cqr->data, sizeof(struct DE_eckd_data));
@@ -1712,6 +1719,13 @@ dasd_3990_update_1B(struct dasd_ccw_req * previous_erp, char *sense)
 		cqr = cqr->refers;
 	}
 
+	if (scsw_is_tm(&cqr->irb.scsw)) {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			      "32 bit sense, action 1B, update,"
+			      " in transport mode - just retry");
+		return previous_erp;
+	}
+
 	/* for imprecise ending just do default erp */
 	if (sense[1] & 0x01) {
 
@@ -2171,7 +2185,7 @@ dasd_3990_erp_control_check(struct dasd_ccw_req *erp)
 {
 	struct dasd_device *device = erp->startdev;
 
-	if (erp->refers->irb.scsw.cmd.cstat & (SCHN_STAT_INTF_CTRL_CHK
+	if (scsw_cstat(&erp->refers->irb.scsw) & (SCHN_STAT_INTF_CTRL_CHK
 					   | SCHN_STAT_CHN_CTRL_CHK)) {
 		DEV_MESSAGE(KERN_DEBUG, device, "%s",
 			    "channel or interface control check");
@@ -2193,21 +2207,23 @@ dasd_3990_erp_control_check(struct dasd_ccw_req *erp)
  *   erp_new		contens was possibly modified
  */
 static struct dasd_ccw_req *
-dasd_3990_erp_inspect(struct dasd_ccw_req * erp)
+dasd_3990_erp_inspect(struct dasd_ccw_req *erp)
 {
 
 	struct dasd_ccw_req *erp_new = NULL;
-	/* sense data are located in the refers record of the */
-	/* already set up new ERP !			      */
-	char *sense = erp->refers->irb.ecw;
+	char *sense;
 
 	/* if this problem occured on an alias retry on base */
 	erp_new = dasd_3990_erp_inspect_alias(erp);
 	if (erp_new)
 		return erp_new;
 
-	/* check if no concurrent sens is available */
-	if (!erp->refers->irb.esw.esw0.erw.cons)
+	/* sense data are located in the refers record of the
+	 * already set up new ERP !
+	 * check if concurrent sens is available
+	 */
+	sense = dasd_get_sense(&erp->refers->irb);
+	if (!sense)
 		erp_new = dasd_3990_erp_control_check(erp);
 	/* distinguish between 24 and 32 byte sense data */
 	else if (sense[27] & DASD_SENSE_BIT_0) {
@@ -2231,7 +2247,11 @@ dasd_3990_erp_inspect(struct dasd_ccw_req * erp)
  * DESCRIPTION
  *   This funtion adds an additional request block (ERP) to the head of
  *   the given cqr (or erp).
- *   This erp is initialized as an default erp (retry TIC)
+ *   For a command mode cqr the erp is initialized as an default erp
+ *   (retry TIC).
+ *   For transport mode we make a copy of the original TCW (points to
+ *   the original TCCB, TIDALs, etc.) but give it a fresh
+ *   TSB so the original sense data will not be changed.
  *
  * PARAMETER
  *   cqr		head of the current ERP-chain (or single cqr if
@@ -2239,17 +2259,27 @@ dasd_3990_erp_inspect(struct dasd_ccw_req * erp)
  * RETURN VALUES
  *   erp		pointer to new ERP-chain head
  */
-static struct dasd_ccw_req *
-dasd_3990_erp_add_erp(struct dasd_ccw_req * cqr)
+static struct dasd_ccw_req *dasd_3990_erp_add_erp(struct dasd_ccw_req *cqr)
 {
 
 	struct dasd_device *device = cqr->startdev;
 	struct ccw1 *ccw;
-
-	/* allocate additional request block */
 	struct dasd_ccw_req *erp;
+	int cplength, datasize;
+	struct tcw *tcw;
+	struct tsb *tsb;
+
+	if (cqr->cpmode == 1) {
+		cplength = 0;
+		datasize = sizeof(struct tcw) + sizeof(struct tsb);
+	} else {
+		cplength = 2;
+		datasize = 0;
+	}
 
-	erp = dasd_alloc_erp_request((char *) &cqr->magic, 2, 0, device);
+	/* allocate additional request block */
+	erp = dasd_alloc_erp_request((char *) &cqr->magic,
+				     cplength, datasize, device);
 	if (IS_ERR(erp)) {
                 if (cqr->retries <= 0) {
 		        DEV_MESSAGE(KERN_ERR, device, "%s",
@@ -2266,13 +2296,24 @@ dasd_3990_erp_add_erp(struct dasd_ccw_req * cqr)
 		return cqr;
 	}
 
-	/* initialize request with default TIC to current ERP/CQR */
-	ccw = erp->cpaddr;
-	ccw->cmd_code = CCW_CMD_NOOP;
-	ccw->flags = CCW_FLAG_CC;
-	ccw++;
-	ccw->cmd_code = CCW_CMD_TIC;
-	ccw->cda      = (long)(cqr->cpaddr);
+	if (cqr->cpmode == 1) {
+		/* make a shallow copy of the original tcw but set new tsb */
+		erp->cpmode = 1;
+		erp->cpaddr = erp->data;
+		tcw = erp->data;
+		tsb = (struct tsb *) &tcw[1];
+		*tcw = *((struct tcw *)cqr->cpaddr);
+		tcw->tsb = (long)tsb;
+	} else {
+		/* initialize request with default TIC to current ERP/CQR */
+		ccw = erp->cpaddr;
+		ccw->cmd_code = CCW_CMD_NOOP;
+		ccw->flags = CCW_FLAG_CC;
+		ccw++;
+		ccw->cmd_code = CCW_CMD_TIC;
+		ccw->cda      = (long)(cqr->cpaddr);
+	}
+
 	erp->function = dasd_3990_erp_add_erp;
 	erp->refers   = cqr;
 	erp->startdev = device;
@@ -2282,7 +2323,6 @@ dasd_3990_erp_add_erp(struct dasd_ccw_req * cqr)
 	erp->expires  = 0;
 	erp->retries  = 256;
 	erp->buildclk = get_clock();
-
 	erp->status = DASD_CQR_FILLED;
 
 	return erp;
@@ -2340,28 +2380,33 @@ dasd_3990_erp_additional_erp(struct dasd_ccw_req * cqr)
  *   match		'boolean' for match found
  *			returns 1 if match found, otherwise 0.
  */
-static int
-dasd_3990_erp_error_match(struct dasd_ccw_req *cqr1, struct dasd_ccw_req *cqr2)
+static int dasd_3990_erp_error_match(struct dasd_ccw_req *cqr1,
+				     struct dasd_ccw_req *cqr2)
 {
+	char *sense1, *sense2;
 
 	if (cqr1->startdev != cqr2->startdev)
 		return 0;
 
-	if (cqr1->irb.esw.esw0.erw.cons != cqr2->irb.esw.esw0.erw.cons)
-		return 0;
+	sense1 = dasd_get_sense(&cqr1->irb);
+	sense2 = dasd_get_sense(&cqr2->irb);
 
-	if ((cqr1->irb.esw.esw0.erw.cons == 0) &&
-	    (cqr2->irb.esw.esw0.erw.cons == 0))	{
-		if ((cqr1->irb.scsw.cmd.cstat & (SCHN_STAT_INTF_CTRL_CHK |
-					     SCHN_STAT_CHN_CTRL_CHK)) ==
-		    (cqr2->irb.scsw.cmd.cstat & (SCHN_STAT_INTF_CTRL_CHK |
-					     SCHN_STAT_CHN_CTRL_CHK)))
+	/* one request has sense data, the other not -> no match, return 0 */
+	if (!sense1 != !sense2)
+		return 0;
+	/* no sense data in both cases -> check cstat for IFCC */
+	if (!sense1 && !sense2)	{
+		if ((scsw_cstat(&cqr1->irb.scsw) & (SCHN_STAT_INTF_CTRL_CHK |
+						    SCHN_STAT_CHN_CTRL_CHK)) ==
+		    (scsw_cstat(&cqr2->irb.scsw) & (SCHN_STAT_INTF_CTRL_CHK |
+						    SCHN_STAT_CHN_CTRL_CHK)))
 			return 1; /* match with ifcc*/
 	}
 	/* check sense data; byte 0-2,25,27 */
-	if (!((memcmp (cqr1->irb.ecw, cqr2->irb.ecw, 3) == 0) &&
-	      (cqr1->irb.ecw[27] == cqr2->irb.ecw[27]) &&
-	      (cqr1->irb.ecw[25] == cqr2->irb.ecw[25]))) {
+	if (!(sense1 && sense2 &&
+	      (memcmp(sense1, sense2, 3) == 0) &&
+	      (sense1[27] == sense2[27]) &&
+	      (sense1[25] == sense2[25]))) {
 
 		return 0;	/* sense doesn't match */
 	}
@@ -2434,7 +2479,7 @@ dasd_3990_erp_further_erp(struct dasd_ccw_req *erp)
 {
 
 	struct dasd_device *device = erp->startdev;
-	char *sense = erp->irb.ecw;
+	char *sense = dasd_get_sense(&erp->irb);
 
 	/* check for 24 byte sense ERP */
 	if ((erp->function == dasd_3990_erp_bus_out) ||
@@ -2449,7 +2494,7 @@ dasd_3990_erp_further_erp(struct dasd_ccw_req *erp)
 		/* prepare erp for retry on different channel path */
 		erp = dasd_3990_erp_action_1(erp);
 
-		if (!(sense[2] & DASD_SENSE_BIT_0)) {
+		if (sense && !(sense[2] & DASD_SENSE_BIT_0)) {
 
 			/* issue a Diagnostic Control command with an
 			 * Inhibit Write subcommand */
@@ -2479,10 +2524,11 @@ dasd_3990_erp_further_erp(struct dasd_ccw_req *erp)
 		}
 
 		/* check for 32 byte sense ERP */
-	} else if ((erp->function == dasd_3990_erp_compound_retry) ||
-		   (erp->function == dasd_3990_erp_compound_path) ||
-		   (erp->function == dasd_3990_erp_compound_code) ||
-		   (erp->function == dasd_3990_erp_compound_config)) {
+	} else if (sense &&
+		   ((erp->function == dasd_3990_erp_compound_retry) ||
+		    (erp->function == dasd_3990_erp_compound_path) ||
+		    (erp->function == dasd_3990_erp_compound_code) ||
+		    (erp->function == dasd_3990_erp_compound_config))) {
 
 		erp = dasd_3990_erp_compound(erp, sense);
 
@@ -2548,18 +2594,19 @@ dasd_3990_erp_handle_match_erp(struct dasd_ccw_req *erp_head,
 
 	if (erp->retries > 0) {
 
-		char *sense = erp->refers->irb.ecw;
+		char *sense = dasd_get_sense(&erp->refers->irb);
 
 		/* check for special retries */
-		if (erp->function == dasd_3990_erp_action_4) {
+		if (sense && erp->function == dasd_3990_erp_action_4) {
 
 			erp = dasd_3990_erp_action_4(erp, sense);
 
-		} else if (erp->function == dasd_3990_erp_action_1B_32) {
+		} else if (sense &&
+			   erp->function == dasd_3990_erp_action_1B_32) {
 
 			erp = dasd_3990_update_1B(erp, sense);
 
-		} else if (erp->function == dasd_3990_erp_int_req) {
+		} else if (sense && erp->function == dasd_3990_erp_int_req) {
 
 			erp = dasd_3990_erp_int_req(erp);
 
@@ -2622,8 +2669,8 @@ dasd_3990_erp_action(struct dasd_ccw_req * cqr)
 	}
 
 	/* double-check if current erp/cqr was successful */
-	if ((cqr->irb.scsw.cmd.cstat == 0x00) &&
-	    (cqr->irb.scsw.cmd.dstat ==
+	if ((scsw_cstat(&cqr->irb.scsw) == 0x00) &&
+	    (scsw_dstat(&cqr->irb.scsw) ==
 	     (DEV_STAT_CHN_END | DEV_STAT_DEV_END))) {
 
 		DEV_MESSAGE(KERN_DEBUG, device,
diff --git a/drivers/s390/block/dasd_alias.c b/drivers/s390/block/dasd_alias.c
index 20676cdef4a5..219bee7bd77c 100644
--- a/drivers/s390/block/dasd_alias.c
+++ b/drivers/s390/block/dasd_alias.c
@@ -646,14 +646,16 @@ static int reset_summary_unit_check(struct alias_lcu *lcu,
 {
 	struct dasd_ccw_req *cqr;
 	int rc = 0;
+	struct ccw1 *ccw;
 
 	cqr = lcu->rsu_cqr;
 	strncpy((char *) &cqr->magic, "ECKD", 4);
 	ASCEBC((char *) &cqr->magic, 4);
-	cqr->cpaddr->cmd_code = DASD_ECKD_CCW_RSCK;
-	cqr->cpaddr->flags = 0 ;
-	cqr->cpaddr->count = 16;
-	cqr->cpaddr->cda = (__u32)(addr_t) cqr->data;
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_RSCK;
+	ccw->flags = 0 ;
+	ccw->count = 16;
+	ccw->cda = (__u32)(addr_t) cqr->data;
 	((char *)cqr->data)[0] = reason;
 
 	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
@@ -855,12 +857,21 @@ void dasd_alias_handle_summary_unit_check(struct dasd_device *device,
 	struct alias_lcu *lcu;
 	char reason;
 	struct dasd_eckd_private *private;
+	char *sense;
 
 	private = (struct dasd_eckd_private *) device->private;
 
-	reason = irb->ecw[8];
-	DEV_MESSAGE(KERN_WARNING, device, "%s %x",
-		    "eckd handle summary unit check: reason", reason);
+	sense = dasd_get_sense(irb);
+	if (sense) {
+		reason = sense[8];
+		DBF_DEV_EVENT(DBF_NOTICE, device, "%s %x",
+			    "eckd handle summary unit check: reason", reason);
+	} else {
+		DBF_DEV_EVENT(DBF_WARNING, device, "%s",
+			    "eckd handle summary unit check:"
+			    " no reason code available");
+		return;
+	}
 
 	lcu = private->lcu;
 	if (!lcu) {
diff --git a/drivers/s390/block/dasd_devmap.c b/drivers/s390/block/dasd_devmap.c
index 34339902efb9..da231a787cee 100644
--- a/drivers/s390/block/dasd_devmap.c
+++ b/drivers/s390/block/dasd_devmap.c
@@ -67,6 +67,8 @@ int dasd_probeonly =  0;	/* is true, when probeonly mode is active */
 int dasd_autodetect = 0;	/* is true, when autodetection is active */
 int dasd_nopav = 0;		/* is true, when PAV is disabled */
 EXPORT_SYMBOL_GPL(dasd_nopav);
+int dasd_nofcx;			/* disable High Performance Ficon */
+EXPORT_SYMBOL_GPL(dasd_nofcx);
 
 /*
  * char *dasd[] is intended to hold the ranges supplied by the dasd= statement
@@ -272,6 +274,11 @@ dasd_parse_keyword( char *parsestring ) {
 		}
 		return residual_str;
 	}
+	if (strncmp("nofcx", parsestring, length) == 0) {
+		dasd_nofcx = 1;
+		MESSAGE(KERN_INFO, "%s", "disable High Performance Ficon");
+		return residual_str;
+	}
 	if (strncmp("fixedbuffers", parsestring, length) == 0) {
 		if (dasd_page_cache)
 			return residual_str;
diff --git a/drivers/s390/block/dasd_eckd.c b/drivers/s390/block/dasd_eckd.c
index 69f93e626fd3..1e4c89b8b304 100644
--- a/drivers/s390/block/dasd_eckd.c
+++ b/drivers/s390/block/dasd_eckd.c
@@ -27,9 +27,12 @@
 #include <asm/uaccess.h>
 #include <asm/cio.h>
 #include <asm/ccwdev.h>
+#include <asm/itcw.h>
 
 #include "dasd_int.h"
 #include "dasd_eckd.h"
+#include "../cio/chsc.h"
+
 
 #ifdef PRINTK_HEADER
 #undef PRINTK_HEADER
@@ -245,7 +248,8 @@ define_extent(struct ccw1 *ccw, struct DE_eckd_data *data, unsigned int trk,
 		rc = check_XRC (ccw, data, device);
 		break;
 	default:
-		DEV_MESSAGE(KERN_ERR, device, "unknown opcode 0x%x", cmd);
+		DBF_DEV_EVENT(DBF_ERR, device,
+			    "PFX LRE unknown opcode 0x%x", cmd);
 		break;
 	}
 
@@ -289,30 +293,145 @@ static int check_XRC_on_prefix(struct PFX_eckd_data *pfxdata,
 		return 0;
 
 	/* switch on System Time Stamp - needed for XRC Support */
-	pfxdata->define_extend.ga_extended |= 0x08; /* 'Time Stamp Valid'   */
-	pfxdata->define_extend.ga_extended |= 0x02; /* 'Extended Parameter' */
+	pfxdata->define_extent.ga_extended |= 0x08; /* 'Time Stamp Valid'   */
+	pfxdata->define_extent.ga_extended |= 0x02; /* 'Extended Parameter' */
 	pfxdata->validity.time_stamp = 1;	    /* 'Time Stamp Valid'   */
 
-	rc = get_sync_clock(&pfxdata->define_extend.ep_sys_time);
+	rc = get_sync_clock(&pfxdata->define_extent.ep_sys_time);
 	/* Ignore return code if sync clock is switched off. */
 	if (rc == -ENOSYS || rc == -EACCES)
 		rc = 0;
 	return rc;
 }
 
-static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
-		  unsigned int trk, unsigned int totrk, int cmd,
-		  struct dasd_device *basedev, struct dasd_device *startdev)
+static void fill_LRE_data(struct LRE_eckd_data *data, unsigned int trk,
+			  unsigned int rec_on_trk, int count, int cmd,
+			  struct dasd_device *device, unsigned int reclen,
+			  unsigned int tlf)
+{
+	struct dasd_eckd_private *private;
+	int sector;
+	int dn, d;
+
+	private = (struct dasd_eckd_private *) device->private;
+
+	memset(data, 0, sizeof(*data));
+	sector = 0;
+	if (rec_on_trk) {
+		switch (private->rdc_data.dev_type) {
+		case 0x3390:
+			dn = ceil_quot(reclen + 6, 232);
+			d = 9 + ceil_quot(reclen + 6 * (dn + 1), 34);
+			sector = (49 + (rec_on_trk - 1) * (10 + d)) / 8;
+			break;
+		case 0x3380:
+			d = 7 + ceil_quot(reclen + 12, 32);
+			sector = (39 + (rec_on_trk - 1) * (8 + d)) / 7;
+			break;
+		}
+	}
+	data->sector = sector;
+	/* note: meaning of count depends on the operation
+	 *	 for record based I/O it's the number of records, but for
+	 *	 track based I/O it's the number of tracks
+	 */
+	data->count = count;
+	switch (cmd) {
+	case DASD_ECKD_CCW_WRITE_HOME_ADDRESS:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x03;
+		break;
+	case DASD_ECKD_CCW_READ_HOME_ADDRESS:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x16;
+		break;
+	case DASD_ECKD_CCW_WRITE_RECORD_ZERO:
+		data->operation.orientation = 0x1;
+		data->operation.operation = 0x03;
+		data->count++;
+		break;
+	case DASD_ECKD_CCW_READ_RECORD_ZERO:
+		data->operation.orientation = 0x3;
+		data->operation.operation = 0x16;
+		data->count++;
+		break;
+	case DASD_ECKD_CCW_WRITE:
+	case DASD_ECKD_CCW_WRITE_MT:
+	case DASD_ECKD_CCW_WRITE_KD:
+	case DASD_ECKD_CCW_WRITE_KD_MT:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x01;
+		break;
+	case DASD_ECKD_CCW_WRITE_CKD:
+	case DASD_ECKD_CCW_WRITE_CKD_MT:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x03;
+		break;
+	case DASD_ECKD_CCW_WRITE_TRACK_DATA:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;	/* not tlf, as one might think */
+		data->operation.operation = 0x3F;
+		data->extended_operation = 0x23;
+		break;
+	case DASD_ECKD_CCW_READ:
+	case DASD_ECKD_CCW_READ_MT:
+	case DASD_ECKD_CCW_READ_KD:
+	case DASD_ECKD_CCW_READ_KD_MT:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x06;
+		break;
+	case DASD_ECKD_CCW_READ_CKD:
+	case DASD_ECKD_CCW_READ_CKD_MT:
+		data->auxiliary.length_valid = 0x1;
+		data->length = reclen;
+		data->operation.operation = 0x16;
+		break;
+	case DASD_ECKD_CCW_READ_COUNT:
+		data->operation.operation = 0x06;
+		break;
+	case DASD_ECKD_CCW_READ_TRACK_DATA:
+		data->auxiliary.length_valid = 0x1;
+		data->length = tlf;
+		data->operation.operation = 0x0C;
+		break;
+	case DASD_ECKD_CCW_ERASE:
+		data->length = reclen;
+		data->auxiliary.length_valid = 0x1;
+		data->operation.operation = 0x0b;
+		break;
+	default:
+		DBF_DEV_EVENT(DBF_ERR, device,
+			    "fill LRE unknown opcode 0x%x", cmd);
+		BUG();
+	}
+	set_ch_t(&data->seek_addr,
+		 trk / private->rdc_data.trk_per_cyl,
+		 trk % private->rdc_data.trk_per_cyl);
+	data->search_arg.cyl = data->seek_addr.cyl;
+	data->search_arg.head = data->seek_addr.head;
+	data->search_arg.record = rec_on_trk;
+}
+
+static int prefix_LRE(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
+		      unsigned int trk, unsigned int totrk, int cmd,
+		      struct dasd_device *basedev, struct dasd_device *startdev,
+		      unsigned char format, unsigned int rec_on_trk, int count,
+		      unsigned int blksize, unsigned int tlf)
 {
 	struct dasd_eckd_private *basepriv, *startpriv;
-	struct DE_eckd_data *data;
+	struct DE_eckd_data *dedata;
+	struct LRE_eckd_data *lredata;
 	u32 begcyl, endcyl;
 	u16 heads, beghead, endhead;
 	int rc = 0;
 
 	basepriv = (struct dasd_eckd_private *) basedev->private;
 	startpriv = (struct dasd_eckd_private *) startdev->private;
-	data = &pfxdata->define_extend;
+	dedata = &pfxdata->define_extent;
+	lredata = &pfxdata->locate_record;
 
 	ccw->cmd_code = DASD_ECKD_CCW_PFX;
 	ccw->flags = 0;
@@ -321,10 +440,16 @@ static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
 
 	memset(pfxdata, 0, sizeof(*pfxdata));
 	/* prefix data */
-	pfxdata->format = 0;
+	if (format > 1) {
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "PFX LRE unknown format 0x%x", format);
+		BUG();
+		return -EINVAL;
+	}
+	pfxdata->format = format;
 	pfxdata->base_address = basepriv->ned->unit_addr;
 	pfxdata->base_lss = basepriv->ned->ID;
-	pfxdata->validity.define_extend = 1;
+	pfxdata->validity.define_extent = 1;
 
 	/* private uid is kept up to date, conf_data may be outdated */
 	if (startpriv->uid.type != UA_BASE_DEVICE) {
@@ -344,42 +469,55 @@ static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
 	case DASD_ECKD_CCW_READ_KD:
 	case DASD_ECKD_CCW_READ_KD_MT:
 	case DASD_ECKD_CCW_READ_COUNT:
-		data->mask.perm = 0x1;
-		data->attributes.operation = basepriv->attrib.operation;
+		dedata->mask.perm = 0x1;
+		dedata->attributes.operation = basepriv->attrib.operation;
+		break;
+	case DASD_ECKD_CCW_READ_TRACK_DATA:
+		dedata->mask.perm = 0x1;
+		dedata->attributes.operation = basepriv->attrib.operation;
+		dedata->blk_size = 0;
 		break;
 	case DASD_ECKD_CCW_WRITE:
 	case DASD_ECKD_CCW_WRITE_MT:
 	case DASD_ECKD_CCW_WRITE_KD:
 	case DASD_ECKD_CCW_WRITE_KD_MT:
-		data->mask.perm = 0x02;
-		data->attributes.operation = basepriv->attrib.operation;
+		dedata->mask.perm = 0x02;
+		dedata->attributes.operation = basepriv->attrib.operation;
 		rc = check_XRC_on_prefix(pfxdata, basedev);
 		break;
 	case DASD_ECKD_CCW_WRITE_CKD:
 	case DASD_ECKD_CCW_WRITE_CKD_MT:
-		data->attributes.operation = DASD_BYPASS_CACHE;
+		dedata->attributes.operation = DASD_BYPASS_CACHE;
 		rc = check_XRC_on_prefix(pfxdata, basedev);
 		break;
 	case DASD_ECKD_CCW_ERASE:
 	case DASD_ECKD_CCW_WRITE_HOME_ADDRESS:
 	case DASD_ECKD_CCW_WRITE_RECORD_ZERO:
-		data->mask.perm = 0x3;
-		data->mask.auth = 0x1;
-		data->attributes.operation = DASD_BYPASS_CACHE;
+		dedata->mask.perm = 0x3;
+		dedata->mask.auth = 0x1;
+		dedata->attributes.operation = DASD_BYPASS_CACHE;
 		rc = check_XRC_on_prefix(pfxdata, basedev);
 		break;
-	default:
-		DEV_MESSAGE(KERN_ERR, basedev, "unknown opcode 0x%x", cmd);
+	case DASD_ECKD_CCW_WRITE_TRACK_DATA:
+		dedata->mask.perm = 0x02;
+		dedata->attributes.operation = basepriv->attrib.operation;
+		dedata->blk_size = blksize;
+		rc = check_XRC_on_prefix(pfxdata, basedev);
 		break;
+	default:
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			    "PFX LRE unknown opcode 0x%x", cmd);
+		BUG();
+		return -EINVAL;
 	}
 
-	data->attributes.mode = 0x3;	/* ECKD */
+	dedata->attributes.mode = 0x3;	/* ECKD */
 
 	if ((basepriv->rdc_data.cu_type == 0x2105 ||
 	     basepriv->rdc_data.cu_type == 0x2107 ||
 	     basepriv->rdc_data.cu_type == 0x1750)
 	    && !(basepriv->uses_cdl && trk < 2))
-		data->ga_extended |= 0x40; /* Regular Data Format Mode */
+		dedata->ga_extended |= 0x40; /* Regular Data Format Mode */
 
 	heads = basepriv->rdc_data.trk_per_cyl;
 	begcyl = trk / heads;
@@ -388,8 +526,8 @@ static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
 	endhead = totrk % heads;
 
 	/* check for sequential prestage - enhance cylinder range */
-	if (data->attributes.operation == DASD_SEQ_PRESTAGE ||
-	    data->attributes.operation == DASD_SEQ_ACCESS) {
+	if (dedata->attributes.operation == DASD_SEQ_PRESTAGE ||
+	    dedata->attributes.operation == DASD_SEQ_ACCESS) {
 
 		if (endcyl + basepriv->attrib.nr_cyl < basepriv->real_cyl)
 			endcyl += basepriv->attrib.nr_cyl;
@@ -397,11 +535,25 @@ static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
 			endcyl = (basepriv->real_cyl - 1);
 	}
 
-	set_ch_t(&data->beg_ext, begcyl, beghead);
-	set_ch_t(&data->end_ext, endcyl, endhead);
+	set_ch_t(&dedata->beg_ext, begcyl, beghead);
+	set_ch_t(&dedata->end_ext, endcyl, endhead);
+
+	if (format == 1) {
+		fill_LRE_data(lredata, trk, rec_on_trk, count, cmd,
+			      basedev, blksize, tlf);
+	}
+
 	return rc;
 }
 
+static int prefix(struct ccw1 *ccw, struct PFX_eckd_data *pfxdata,
+		  unsigned int trk, unsigned int totrk, int cmd,
+		  struct dasd_device *basedev, struct dasd_device *startdev)
+{
+	return prefix_LRE(ccw, pfxdata, trk, totrk, cmd, basedev, startdev,
+			  0, 0, 0, 0, 0);
+}
+
 static void
 locate_record(struct ccw1 *ccw, struct LO_eckd_data *data, unsigned int trk,
 	      unsigned int rec_on_trk, int no_rec, int cmd,
@@ -845,7 +997,8 @@ static int dasd_eckd_read_features(struct dasd_device *device)
 /*
  * Build CP for Perform Subsystem Function - SSC.
  */
-static struct dasd_ccw_req *dasd_eckd_build_psf_ssc(struct dasd_device *device)
+static struct dasd_ccw_req *dasd_eckd_build_psf_ssc(struct dasd_device *device,
+						    int enable_pav)
 {
 	struct dasd_ccw_req *cqr;
 	struct dasd_psf_ssc_data *psf_ssc_data;
@@ -862,9 +1015,11 @@ static struct dasd_ccw_req *dasd_eckd_build_psf_ssc(struct dasd_device *device)
 	}
 	psf_ssc_data = (struct dasd_psf_ssc_data *)cqr->data;
 	psf_ssc_data->order = PSF_ORDER_SSC;
-	psf_ssc_data->suborder = 0x88;
-	psf_ssc_data->reserved[0] = 0x88;
-
+	psf_ssc_data->suborder = 0x40;
+	if (enable_pav) {
+		psf_ssc_data->suborder |= 0x88;
+		psf_ssc_data->reserved[0] = 0x88;
+	}
 	ccw = cqr->cpaddr;
 	ccw->cmd_code = DASD_ECKD_CCW_PSF;
 	ccw->cda = (__u32)(addr_t)psf_ssc_data;
@@ -885,12 +1040,12 @@ static struct dasd_ccw_req *dasd_eckd_build_psf_ssc(struct dasd_device *device)
  * call might change behaviour of DASD devices.
  */
 static int
-dasd_eckd_psf_ssc(struct dasd_device *device)
+dasd_eckd_psf_ssc(struct dasd_device *device, int enable_pav)
 {
 	struct dasd_ccw_req *cqr;
 	int rc;
 
-	cqr = dasd_eckd_build_psf_ssc(device);
+	cqr = dasd_eckd_build_psf_ssc(device, enable_pav);
 	if (IS_ERR(cqr))
 		return PTR_ERR(cqr);
 
@@ -909,12 +1064,13 @@ static int dasd_eckd_validate_server(struct dasd_device *device)
 {
 	int rc;
 	struct dasd_eckd_private *private;
+	int enable_pav;
 
-	/* Currently PAV is the only reason to 'validate' server on LPAR */
 	if (dasd_nopav || MACHINE_IS_VM)
-		return 0;
-
-	rc = dasd_eckd_psf_ssc(device);
+		enable_pav = 0;
+	else
+		enable_pav = 1;
+	rc = dasd_eckd_psf_ssc(device, enable_pav);
 	/* may be requested feature is not available on server,
 	 * therefore just report error and go ahead */
 	private = (struct dasd_eckd_private *) device->private;
@@ -1504,40 +1660,41 @@ static void dasd_eckd_handle_unsolicited_interrupt(struct dasd_device *device,
 						   struct irb *irb)
 {
 	char mask;
+	char *sense = NULL;
 
 	/* first of all check for state change pending interrupt */
 	mask = DEV_STAT_ATTENTION | DEV_STAT_DEV_END | DEV_STAT_UNIT_EXCEP;
-	if ((irb->scsw.cmd.dstat & mask) == mask) {
+	if ((scsw_dstat(&irb->scsw) & mask) == mask) {
 		dasd_generic_handle_state_change(device);
 		return;
 	}
 
 	/* summary unit check */
-	if ((irb->scsw.cmd.dstat & DEV_STAT_UNIT_CHECK) &&
+	if ((scsw_dstat(&irb->scsw) & DEV_STAT_UNIT_CHECK) &&
 	    (irb->ecw[7] == 0x0D)) {
 		dasd_alias_handle_summary_unit_check(device, irb);
 		return;
 	}
 
-
+	sense = dasd_get_sense(irb);
 	/* service information message SIM */
-	if (irb->esw.esw0.erw.cons && !(irb->ecw[27] & DASD_SENSE_BIT_0) &&
-	    ((irb->ecw[6] & DASD_SIM_SENSE) == DASD_SIM_SENSE)) {
-		dasd_3990_erp_handle_sim(device, irb->ecw);
+	if (sense && !(sense[27] & DASD_SENSE_BIT_0) &&
+	    ((sense[6] & DASD_SIM_SENSE) == DASD_SIM_SENSE)) {
+		dasd_3990_erp_handle_sim(device, sense);
 		dasd_schedule_device_bh(device);
 		return;
 	}
 
-	if ((irb->scsw.cmd.cc == 1) &&
-	    (irb->scsw.cmd.fctl & SCSW_FCTL_START_FUNC) &&
-	    (irb->scsw.cmd.actl & SCSW_ACTL_START_PEND) &&
-	    (irb->scsw.cmd.stctl & SCSW_STCTL_STATUS_PEND)) {
+	if ((scsw_cc(&irb->scsw) == 1) &&
+	    (scsw_fctl(&irb->scsw) & SCSW_FCTL_START_FUNC) &&
+	    (scsw_actl(&irb->scsw) & SCSW_ACTL_START_PEND) &&
+	    (scsw_stctl(&irb->scsw) & SCSW_STCTL_STATUS_PEND)) {
 		/* fake irb do nothing, they are handled elsewhere */
 		dasd_schedule_device_bh(device);
 		return;
 	}
 
-	if (!(irb->esw.esw0.erw.cons)) {
+	if (!sense) {
 		/* just report other unsolicited interrupts */
 		DEV_MESSAGE(KERN_ERR, device, "%s",
 			    "unsolicited interrupt received");
@@ -1552,9 +1709,19 @@ static void dasd_eckd_handle_unsolicited_interrupt(struct dasd_device *device,
 	return;
 };
 
-static struct dasd_ccw_req *dasd_eckd_build_cp(struct dasd_device *startdev,
+
+static struct dasd_ccw_req *dasd_eckd_build_cp_cmd_single(
+					       struct dasd_device *startdev,
 					       struct dasd_block *block,
-					       struct request *req)
+					       struct request *req,
+					       sector_t first_rec,
+					       sector_t last_rec,
+					       sector_t first_trk,
+					       sector_t last_trk,
+					       unsigned int first_offs,
+					       unsigned int last_offs,
+					       unsigned int blk_per_trk,
+					       unsigned int blksize)
 {
 	struct dasd_eckd_private *private;
 	unsigned long *idaws;
@@ -1564,11 +1731,9 @@ static struct dasd_ccw_req *dasd_eckd_build_cp(struct dasd_device *startdev,
 	struct req_iterator iter;
 	struct bio_vec *bv;
 	char *dst;
-	unsigned int blksize, blk_per_trk, off;
+	unsigned int off;
 	int count, cidaw, cplength, datasize;
-	sector_t recid, first_rec, last_rec;
-	sector_t first_trk, last_trk;
-	unsigned int first_offs, last_offs;
+	sector_t recid;
 	unsigned char cmd, rcmd;
 	int use_prefix;
 	struct dasd_device *basedev;
@@ -1581,15 +1746,7 @@ static struct dasd_ccw_req *dasd_eckd_build_cp(struct dasd_device *startdev,
 		cmd = DASD_ECKD_CCW_WRITE_MT;
 	else
 		return ERR_PTR(-EINVAL);
-	/* Calculate number of blocks/records per track. */
-	blksize = block->bp_block;
-	blk_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
-	/* Calculate record id of first and last block. */
-	first_rec = first_trk = req->sector >> block->s2b_shift;
-	first_offs = sector_div(first_trk, blk_per_trk);
-	last_rec = last_trk =
-		(req->sector + req->nr_sectors - 1) >> block->s2b_shift;
-	last_offs = sector_div(last_trk, blk_per_trk);
+
 	/* Check struct bio and count the number of blocks for the request. */
 	count = 0;
 	cidaw = 0;
@@ -1739,6 +1896,497 @@ static struct dasd_ccw_req *dasd_eckd_build_cp(struct dasd_device *startdev,
 	return cqr;
 }
 
+static struct dasd_ccw_req *dasd_eckd_build_cp_cmd_track(
+					       struct dasd_device *startdev,
+					       struct dasd_block *block,
+					       struct request *req,
+					       sector_t first_rec,
+					       sector_t last_rec,
+					       sector_t first_trk,
+					       sector_t last_trk,
+					       unsigned int first_offs,
+					       unsigned int last_offs,
+					       unsigned int blk_per_trk,
+					       unsigned int blksize)
+{
+	struct dasd_eckd_private *private;
+	unsigned long *idaws;
+	struct dasd_ccw_req *cqr;
+	struct ccw1 *ccw;
+	struct req_iterator iter;
+	struct bio_vec *bv;
+	char *dst, *idaw_dst;
+	unsigned int cidaw, cplength, datasize;
+	unsigned int tlf;
+	sector_t recid;
+	unsigned char cmd;
+	struct dasd_device *basedev;
+	unsigned int trkcount, count, count_to_trk_end;
+	unsigned int idaw_len, seg_len, part_len, len_to_track_end;
+	unsigned char new_track, end_idaw;
+	sector_t trkid;
+	unsigned int recoffs;
+
+	basedev = block->base;
+	private = (struct dasd_eckd_private *) basedev->private;
+	if (rq_data_dir(req) == READ)
+		cmd = DASD_ECKD_CCW_READ_TRACK_DATA;
+	else if (rq_data_dir(req) == WRITE)
+		cmd = DASD_ECKD_CCW_WRITE_TRACK_DATA;
+	else
+		return ERR_PTR(-EINVAL);
+
+	/* Track based I/O needs IDAWs for each page, and not just for
+	 * 64 bit addresses. We need additional idals for pages
+	 * that get filled from two tracks, so we use the number
+	 * of records as upper limit.
+	 */
+	cidaw = last_rec - first_rec + 1;
+	trkcount = last_trk - first_trk + 1;
+
+	/* 1x prefix + one read/write ccw per track */
+	cplength = 1 + trkcount;
+
+	/* on 31-bit we need space for two 32 bit addresses per page
+	 * on 64-bit one 64 bit address
+	 */
+	datasize = sizeof(struct PFX_eckd_data) +
+		cidaw * sizeof(unsigned long long);
+
+	/* Allocate the ccw request. */
+	cqr = dasd_smalloc_request(dasd_eckd_discipline.name,
+				   cplength, datasize, startdev);
+	if (IS_ERR(cqr))
+		return cqr;
+	ccw = cqr->cpaddr;
+	/* transfer length factor: how many bytes to read from the last track */
+	if (first_trk == last_trk)
+		tlf = last_offs - first_offs + 1;
+	else
+		tlf = last_offs + 1;
+	tlf *= blksize;
+
+	if (prefix_LRE(ccw++, cqr->data, first_trk,
+		       last_trk, cmd, basedev, startdev,
+		       1 /* format */, first_offs + 1,
+		       trkcount, blksize,
+		       tlf) == -EAGAIN) {
+		/* Clock not in sync and XRC is enabled.
+		 * Try again later.
+		 */
+		dasd_sfree_request(cqr, startdev);
+		return ERR_PTR(-EAGAIN);
+	}
+
+	/*
+	 * The translation of request into ccw programs must meet the
+	 * following conditions:
+	 * - all idaws but the first and the last must address full pages
+	 *   (or 2K blocks on 31-bit)
+	 * - the scope of a ccw and it's idal ends with the track boundaries
+	 */
+	idaws = (unsigned long *) (cqr->data + sizeof(struct PFX_eckd_data));
+	recid = first_rec;
+	new_track = 1;
+	end_idaw = 0;
+	len_to_track_end = 0;
+	idaw_dst = 0;
+	idaw_len = 0;
+	rq_for_each_segment(bv, req, iter) {
+		dst = page_address(bv->bv_page) + bv->bv_offset;
+		seg_len = bv->bv_len;
+		while (seg_len) {
+			if (new_track) {
+				trkid = recid;
+				recoffs = sector_div(trkid, blk_per_trk);
+				count_to_trk_end = blk_per_trk - recoffs;
+				count = min((last_rec - recid + 1),
+					    (sector_t)count_to_trk_end);
+				len_to_track_end = count * blksize;
+				ccw[-1].flags |= CCW_FLAG_CC;
+				ccw->cmd_code = cmd;
+				ccw->count = len_to_track_end;
+				ccw->cda = (__u32)(addr_t)idaws;
+				ccw->flags = CCW_FLAG_IDA;
+				ccw++;
+				recid += count;
+				new_track = 0;
+			}
+			/* If we start a new idaw, everything is fine and the
+			 * start of the new idaw is the start of this segment.
+			 * If we continue an idaw, we must make sure that the
+			 * current segment begins where the so far accumulated
+			 * idaw ends
+			 */
+			if (!idaw_dst)
+				idaw_dst = dst;
+			if ((idaw_dst + idaw_len) != dst) {
+				dasd_sfree_request(cqr, startdev);
+				return ERR_PTR(-ERANGE);
+			}
+			part_len = min(seg_len, len_to_track_end);
+			seg_len -= part_len;
+			dst += part_len;
+			idaw_len += part_len;
+			len_to_track_end -= part_len;
+			/* collected memory area ends on an IDA_BLOCK border,
+			 * -> create an idaw
+			 * idal_create_words will handle cases where idaw_len
+			 * is larger then IDA_BLOCK_SIZE
+			 */
+			if (!(__pa(idaw_dst + idaw_len) & (IDA_BLOCK_SIZE-1)))
+				end_idaw = 1;
+			/* We also need to end the idaw at track end */
+			if (!len_to_track_end) {
+				new_track = 1;
+				end_idaw = 1;
+			}
+			if (end_idaw) {
+				idaws = idal_create_words(idaws, idaw_dst,
+							  idaw_len);
+				idaw_dst = 0;
+				idaw_len = 0;
+				end_idaw = 0;
+			}
+		}
+	}
+
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->block = block;
+	cqr->expires = 5 * 60 * HZ;	/* 5 minutes */
+	cqr->lpm = private->path_data.ppm;
+	cqr->retries = 256;
+	cqr->buildclk = get_clock();
+	cqr->status = DASD_CQR_FILLED;
+	return cqr;
+}
+
+static int prepare_itcw(struct itcw *itcw,
+			unsigned int trk, unsigned int totrk, int cmd,
+			struct dasd_device *basedev,
+			struct dasd_device *startdev,
+			unsigned int rec_on_trk, int count,
+			unsigned int blksize,
+			unsigned int total_data_size,
+			unsigned int tlf,
+			unsigned int blk_per_trk)
+{
+	struct PFX_eckd_data pfxdata;
+	struct dasd_eckd_private *basepriv, *startpriv;
+	struct DE_eckd_data *dedata;
+	struct LRE_eckd_data *lredata;
+	struct dcw *dcw;
+
+	u32 begcyl, endcyl;
+	u16 heads, beghead, endhead;
+	u8 pfx_cmd;
+
+	int rc = 0;
+	int sector = 0;
+	int dn, d;
+
+
+	/* setup prefix data */
+	basepriv = (struct dasd_eckd_private *) basedev->private;
+	startpriv = (struct dasd_eckd_private *) startdev->private;
+	dedata = &pfxdata.define_extent;
+	lredata = &pfxdata.locate_record;
+
+	memset(&pfxdata, 0, sizeof(pfxdata));
+	pfxdata.format = 1; /* PFX with LRE */
+	pfxdata.base_address = basepriv->ned->unit_addr;
+	pfxdata.base_lss = basepriv->ned->ID;
+	pfxdata.validity.define_extent = 1;
+
+	/* private uid is kept up to date, conf_data may be outdated */
+	if (startpriv->uid.type != UA_BASE_DEVICE) {
+		pfxdata.validity.verify_base = 1;
+		if (startpriv->uid.type == UA_HYPER_PAV_ALIAS)
+			pfxdata.validity.hyper_pav = 1;
+	}
+
+	switch (cmd) {
+	case DASD_ECKD_CCW_READ_TRACK_DATA:
+		dedata->mask.perm = 0x1;
+		dedata->attributes.operation = basepriv->attrib.operation;
+		dedata->blk_size = blksize;
+		dedata->ga_extended |= 0x42;
+		lredata->operation.orientation = 0x0;
+		lredata->operation.operation = 0x0C;
+		lredata->auxiliary.check_bytes = 0x01;
+		pfx_cmd = DASD_ECKD_CCW_PFX_READ;
+		break;
+	case DASD_ECKD_CCW_WRITE_TRACK_DATA:
+		dedata->mask.perm = 0x02;
+		dedata->attributes.operation = basepriv->attrib.operation;
+		dedata->blk_size = blksize;
+		rc = check_XRC_on_prefix(&pfxdata, basedev);
+		dedata->ga_extended |= 0x42;
+		lredata->operation.orientation = 0x0;
+		lredata->operation.operation = 0x3F;
+		lredata->extended_operation = 0x23;
+		lredata->auxiliary.check_bytes = 0x2;
+		pfx_cmd = DASD_ECKD_CCW_PFX;
+		break;
+	default:
+		DBF_DEV_EVENT(DBF_ERR, basedev,
+			      "prepare itcw, unknown opcode 0x%x", cmd);
+		BUG();
+		break;
+	}
+	if (rc)
+		return rc;
+
+	dedata->attributes.mode = 0x3;	/* ECKD */
+
+	heads = basepriv->rdc_data.trk_per_cyl;
+	begcyl = trk / heads;
+	beghead = trk % heads;
+	endcyl = totrk / heads;
+	endhead = totrk % heads;
+
+	/* check for sequential prestage - enhance cylinder range */
+	if (dedata->attributes.operation == DASD_SEQ_PRESTAGE ||
+	    dedata->attributes.operation == DASD_SEQ_ACCESS) {
+
+		if (endcyl + basepriv->attrib.nr_cyl < basepriv->real_cyl)
+			endcyl += basepriv->attrib.nr_cyl;
+		else
+			endcyl = (basepriv->real_cyl - 1);
+	}
+
+	set_ch_t(&dedata->beg_ext, begcyl, beghead);
+	set_ch_t(&dedata->end_ext, endcyl, endhead);
+
+	dedata->ep_format = 0x20; /* records per track is valid */
+	dedata->ep_rec_per_track = blk_per_trk;
+
+	if (rec_on_trk) {
+		switch (basepriv->rdc_data.dev_type) {
+		case 0x3390:
+			dn = ceil_quot(blksize + 6, 232);
+			d = 9 + ceil_quot(blksize + 6 * (dn + 1), 34);
+			sector = (49 + (rec_on_trk - 1) * (10 + d)) / 8;
+			break;
+		case 0x3380:
+			d = 7 + ceil_quot(blksize + 12, 32);
+			sector = (39 + (rec_on_trk - 1) * (8 + d)) / 7;
+			break;
+		}
+	}
+
+	lredata->auxiliary.length_valid = 1;
+	lredata->auxiliary.length_scope = 1;
+	lredata->auxiliary.imbedded_ccw_valid = 1;
+	lredata->length = tlf;
+	lredata->imbedded_ccw = cmd;
+	lredata->count = count;
+	lredata->sector = sector;
+	set_ch_t(&lredata->seek_addr, begcyl, beghead);
+	lredata->search_arg.cyl = lredata->seek_addr.cyl;
+	lredata->search_arg.head = lredata->seek_addr.head;
+	lredata->search_arg.record = rec_on_trk;
+
+	dcw = itcw_add_dcw(itcw, pfx_cmd, 0,
+		     &pfxdata, sizeof(pfxdata), total_data_size);
+
+	return rc;
+}
+
+static struct dasd_ccw_req *dasd_eckd_build_cp_tpm_track(
+					       struct dasd_device *startdev,
+					       struct dasd_block *block,
+					       struct request *req,
+					       sector_t first_rec,
+					       sector_t last_rec,
+					       sector_t first_trk,
+					       sector_t last_trk,
+					       unsigned int first_offs,
+					       unsigned int last_offs,
+					       unsigned int blk_per_trk,
+					       unsigned int blksize)
+{
+	struct dasd_eckd_private *private;
+	struct dasd_ccw_req *cqr;
+	struct req_iterator iter;
+	struct bio_vec *bv;
+	char *dst;
+	unsigned int trkcount, ctidaw;
+	unsigned char cmd;
+	struct dasd_device *basedev;
+	unsigned int tlf;
+	struct itcw *itcw;
+	struct tidaw *last_tidaw = NULL;
+	int itcw_op;
+	size_t itcw_size;
+
+	basedev = block->base;
+	private = (struct dasd_eckd_private *) basedev->private;
+	if (rq_data_dir(req) == READ) {
+		cmd = DASD_ECKD_CCW_READ_TRACK_DATA;
+		itcw_op = ITCW_OP_READ;
+	} else if (rq_data_dir(req) == WRITE) {
+		cmd = DASD_ECKD_CCW_WRITE_TRACK_DATA;
+		itcw_op = ITCW_OP_WRITE;
+	} else
+		return ERR_PTR(-EINVAL);
+
+	/* trackbased I/O needs address all memory via TIDAWs,
+	 * not just for 64 bit addresses. This allows us to map
+	 * each segment directly to one tidaw.
+	 */
+	trkcount = last_trk - first_trk + 1;
+	ctidaw = 0;
+	rq_for_each_segment(bv, req, iter) {
+		++ctidaw;
+	}
+
+	/* Allocate the ccw request. */
+	itcw_size = itcw_calc_size(0, ctidaw, 0);
+	cqr = dasd_smalloc_request(dasd_eckd_discipline.name,
+				   0, itcw_size, startdev);
+	if (IS_ERR(cqr))
+		return cqr;
+
+	cqr->cpmode = 1;
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->block = block;
+	cqr->expires = 100*HZ;
+	cqr->buildclk = get_clock();
+	cqr->status = DASD_CQR_FILLED;
+	cqr->retries = 10;
+
+	/* transfer length factor: how many bytes to read from the last track */
+	if (first_trk == last_trk)
+		tlf = last_offs - first_offs + 1;
+	else
+		tlf = last_offs + 1;
+	tlf *= blksize;
+
+	itcw = itcw_init(cqr->data, itcw_size, itcw_op, 0, ctidaw, 0);
+	cqr->cpaddr = itcw_get_tcw(itcw);
+
+	if (prepare_itcw(itcw, first_trk, last_trk,
+			 cmd, basedev, startdev,
+			 first_offs + 1,
+			 trkcount, blksize,
+			 (last_rec - first_rec + 1) * blksize,
+			 tlf, blk_per_trk) == -EAGAIN) {
+		/* Clock not in sync and XRC is enabled.
+		 * Try again later.
+		 */
+		dasd_sfree_request(cqr, startdev);
+		return ERR_PTR(-EAGAIN);
+	}
+
+	/*
+	 * A tidaw can address 4k of memory, but must not cross page boundaries
+	 * We can let the block layer handle this by setting
+	 * blk_queue_segment_boundary to page boundaries and
+	 * blk_max_segment_size to page size when setting up the request queue.
+	 */
+	rq_for_each_segment(bv, req, iter) {
+		dst = page_address(bv->bv_page) + bv->bv_offset;
+		last_tidaw = itcw_add_tidaw(itcw, 0x00, dst, bv->bv_len);
+		if (IS_ERR(last_tidaw))
+			return (struct dasd_ccw_req *)last_tidaw;
+	}
+
+	last_tidaw->flags |= 0x80;
+	itcw_finalize(itcw);
+
+	if (blk_noretry_request(req) ||
+	    block->base->features & DASD_FEATURE_FAILFAST)
+		set_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags);
+	cqr->startdev = startdev;
+	cqr->memdev = startdev;
+	cqr->block = block;
+	cqr->expires = 5 * 60 * HZ;	/* 5 minutes */
+	cqr->lpm = private->path_data.ppm;
+	cqr->retries = 256;
+	cqr->buildclk = get_clock();
+	cqr->status = DASD_CQR_FILLED;
+	return cqr;
+}
+
+static struct dasd_ccw_req *dasd_eckd_build_cp(struct dasd_device *startdev,
+					       struct dasd_block *block,
+					       struct request *req)
+{
+	int tpm, cmdrtd, cmdwtd;
+	int use_prefix;
+
+	struct dasd_eckd_private *private;
+	int fcx_in_css, fcx_in_gneq, fcx_in_features;
+	struct dasd_device *basedev;
+	sector_t first_rec, last_rec;
+	sector_t first_trk, last_trk;
+	unsigned int first_offs, last_offs;
+	unsigned int blk_per_trk, blksize;
+	int cdlspecial;
+	struct dasd_ccw_req *cqr;
+
+	basedev = block->base;
+	private = (struct dasd_eckd_private *) basedev->private;
+
+	/* Calculate number of blocks/records per track. */
+	blksize = block->bp_block;
+	blk_per_trk = recs_per_track(&private->rdc_data, 0, blksize);
+	/* Calculate record id of first and last block. */
+	first_rec = first_trk = req->sector >> block->s2b_shift;
+	first_offs = sector_div(first_trk, blk_per_trk);
+	last_rec = last_trk =
+		(req->sector + req->nr_sectors - 1) >> block->s2b_shift;
+	last_offs = sector_div(last_trk, blk_per_trk);
+	cdlspecial = (private->uses_cdl && first_rec < 2*blk_per_trk);
+
+	/* is transport mode supported ? */
+	fcx_in_css = css_general_characteristics.fcx;
+	fcx_in_gneq = private->gneq->reserved2[7] & 0x04;
+	fcx_in_features = private->features.feature[40] & 0x80;
+	tpm = fcx_in_css && fcx_in_gneq && fcx_in_features;
+
+	/* is read track data and write track data in command mode supported? */
+	cmdrtd = private->features.feature[9] & 0x20;
+	cmdwtd = private->features.feature[12] & 0x40;
+	use_prefix = private->features.feature[8] & 0x01;
+
+	cqr = NULL;
+	if (cdlspecial || dasd_page_cache) {
+		/* do nothing, just fall through to the cmd mode single case */
+	} else if (!dasd_nofcx && tpm && (first_trk == last_trk)) {
+		cqr = dasd_eckd_build_cp_tpm_track(startdev, block, req,
+						    first_rec, last_rec,
+						    first_trk, last_trk,
+						    first_offs, last_offs,
+						    blk_per_trk, blksize);
+		if (IS_ERR(cqr) && PTR_ERR(cqr) != -EAGAIN)
+			cqr = NULL;
+	} else if (use_prefix &&
+		   (((rq_data_dir(req) == READ) && cmdrtd) ||
+		    ((rq_data_dir(req) == WRITE) && cmdwtd))) {
+		cqr = dasd_eckd_build_cp_cmd_track(startdev, block, req,
+						   first_rec, last_rec,
+						   first_trk, last_trk,
+						   first_offs, last_offs,
+						   blk_per_trk, blksize);
+		if (IS_ERR(cqr) && PTR_ERR(cqr) != -EAGAIN)
+			cqr = NULL;
+	}
+	if (!cqr)
+		cqr = dasd_eckd_build_cp_cmd_single(startdev, block, req,
+						    first_rec, last_rec,
+						    first_trk, last_trk,
+						    first_offs, last_offs,
+						    blk_per_trk, blksize);
+	return cqr;
+}
+
 static int
 dasd_eckd_free_cp(struct dasd_ccw_req *cqr, struct request *req)
 {
@@ -1792,7 +2440,7 @@ out:
 }
 
 /*
- * Modify ccw chain in cqr so it can be started on a base device.
+ * Modify ccw/tcw in cqr so it can be started on a base device.
  *
  * Note that this is not enough to restart the cqr!
  * Either reset cqr->startdev as well (summary unit check handling)
@@ -1802,13 +2450,24 @@ void dasd_eckd_reset_ccw_to_base_io(struct dasd_ccw_req *cqr)
 {
 	struct ccw1 *ccw;
 	struct PFX_eckd_data *pfxdata;
-
-	ccw = cqr->cpaddr;
-	pfxdata = cqr->data;
-
-	if (ccw->cmd_code == DASD_ECKD_CCW_PFX) {
+	struct tcw *tcw;
+	struct tccb *tccb;
+	struct dcw *dcw;
+
+	if (cqr->cpmode == 1) {
+		tcw = cqr->cpaddr;
+		tccb = tcw_get_tccb(tcw);
+		dcw = (struct dcw *)&tccb->tca[0];
+		pfxdata = (struct PFX_eckd_data *)&dcw->cd[0];
 		pfxdata->validity.verify_base = 0;
 		pfxdata->validity.hyper_pav = 0;
+	} else {
+		ccw = cqr->cpaddr;
+		pfxdata = cqr->data;
+		if (ccw->cmd_code == DASD_ECKD_CCW_PFX) {
+			pfxdata->validity.verify_base = 0;
+			pfxdata->validity.hyper_pav = 0;
+		}
 	}
 }
 
@@ -1886,6 +2545,7 @@ dasd_eckd_release(struct dasd_device *device)
 {
 	struct dasd_ccw_req *cqr;
 	int rc;
+	struct ccw1 *ccw;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
@@ -1897,10 +2557,11 @@ dasd_eckd_release(struct dasd_device *device)
 			    "Could not allocate initialization request");
 		return PTR_ERR(cqr);
 	}
-	cqr->cpaddr->cmd_code = DASD_ECKD_CCW_RELEASE;
-        cqr->cpaddr->flags |= CCW_FLAG_SLI;
-        cqr->cpaddr->count = 32;
-	cqr->cpaddr->cda = (__u32)(addr_t) cqr->data;
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_RELEASE;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = 32;
+	ccw->cda = (__u32)(addr_t) cqr->data;
 	cqr->startdev = device;
 	cqr->memdev = device;
 	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
@@ -1927,6 +2588,7 @@ dasd_eckd_reserve(struct dasd_device *device)
 {
 	struct dasd_ccw_req *cqr;
 	int rc;
+	struct ccw1 *ccw;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
@@ -1938,10 +2600,11 @@ dasd_eckd_reserve(struct dasd_device *device)
 			    "Could not allocate initialization request");
 		return PTR_ERR(cqr);
 	}
-	cqr->cpaddr->cmd_code = DASD_ECKD_CCW_RESERVE;
-        cqr->cpaddr->flags |= CCW_FLAG_SLI;
-        cqr->cpaddr->count = 32;
-	cqr->cpaddr->cda = (__u32)(addr_t) cqr->data;
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_RESERVE;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = 32;
+	ccw->cda = (__u32)(addr_t) cqr->data;
 	cqr->startdev = device;
 	cqr->memdev = device;
 	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
@@ -1967,6 +2630,7 @@ dasd_eckd_steal_lock(struct dasd_device *device)
 {
 	struct dasd_ccw_req *cqr;
 	int rc;
+	struct ccw1 *ccw;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
@@ -1978,10 +2642,11 @@ dasd_eckd_steal_lock(struct dasd_device *device)
 			    "Could not allocate initialization request");
 		return PTR_ERR(cqr);
 	}
-	cqr->cpaddr->cmd_code = DASD_ECKD_CCW_SLCK;
-        cqr->cpaddr->flags |= CCW_FLAG_SLI;
-        cqr->cpaddr->count = 32;
-	cqr->cpaddr->cda = (__u32)(addr_t) cqr->data;
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_SLCK;
+	ccw->flags |= CCW_FLAG_SLI;
+	ccw->count = 32;
+	ccw->cda = (__u32)(addr_t) cqr->data;
 	cqr->startdev = device;
 	cqr->memdev = device;
 	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
@@ -2271,7 +2936,7 @@ dasd_eckd_dump_ccw_range(struct ccw1 *from, struct ccw1 *to, char *page)
  * Print sense data and related channel program.
  * Parts are printed because printk buffer is only 1024 bytes.
  */
-static void dasd_eckd_dump_sense(struct dasd_device *device,
+static void dasd_eckd_dump_sense_ccw(struct dasd_device *device,
 				 struct dasd_ccw_req *req, struct irb *irb)
 {
 	char *page;
@@ -2290,7 +2955,7 @@ static void dasd_eckd_dump_sense(struct dasd_device *device,
 		      dev_name(&device->cdev->dev));
 	len += sprintf(page + len, KERN_ERR PRINTK_HEADER
 		       " in req: %p CS: 0x%02X DS: 0x%02X\n", req,
-		       irb->scsw.cmd.cstat, irb->scsw.cmd.dstat);
+		       scsw_cstat(&irb->scsw), scsw_dstat(&irb->scsw));
 	len += sprintf(page + len, KERN_ERR PRINTK_HEADER
 		       " device %s: Failing CCW: %p\n",
 		       dev_name(&device->cdev->dev),
@@ -2366,6 +3031,147 @@ static void dasd_eckd_dump_sense(struct dasd_device *device,
 	free_page((unsigned long) page);
 }
 
+
+/*
+ * Print sense data from a tcw.
+ */
+static void dasd_eckd_dump_sense_tcw(struct dasd_device *device,
+				 struct dasd_ccw_req *req, struct irb *irb)
+{
+	char *page;
+	int len, sl, sct, residual;
+
+	struct tsb *tsb;
+	u8 *sense;
+
+
+	page = (char *) get_zeroed_page(GFP_ATOMIC);
+	if (page == NULL) {
+		DEV_MESSAGE(KERN_ERR, device, " %s",
+			    "No memory to dump sense data");
+		return;
+	}
+	/* dump the sense data */
+	len = sprintf(page,  KERN_ERR PRINTK_HEADER
+		      " I/O status report for device %s:\n",
+		      dev_name(&device->cdev->dev));
+	len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+		       " in req: %p CS: 0x%02X DS: 0x%02X "
+		       "fcxs: 0x%02X schxs: 0x%02X\n", req,
+		       scsw_cstat(&irb->scsw), scsw_dstat(&irb->scsw),
+		       irb->scsw.tm.fcxs, irb->scsw.tm.schxs);
+	len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+		       " device %s: Failing TCW: %p\n",
+		       dev_name(&device->cdev->dev),
+		       (void *) (addr_t) irb->scsw.tm.tcw);
+
+	tsb = NULL;
+	sense = NULL;
+	if (irb->scsw.tm.tcw)
+		tsb = tcw_get_tsb(
+			(struct tcw *)(unsigned long)irb->scsw.tm.tcw);
+
+	if (tsb && (irb->scsw.tm.fcxs == 0x01)) {
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->length %d\n", tsb->length);
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->flags %x\n", tsb->flags);
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->dcw_offset %d\n", tsb->dcw_offset);
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->count %d\n", tsb->count);
+		residual = tsb->count - 28;
+		len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " residual %d\n", residual);
+
+		switch (tsb->flags & 0x07) {
+		case 1:	/* tsa_iostat */
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->tsa.iostat.dev_time %d\n",
+				       tsb->tsa.iostat.dev_time);
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->tsa.iostat.def_time %d\n",
+				       tsb->tsa.iostat.def_time);
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->tsa.iostat.queue_time %d\n",
+				       tsb->tsa.iostat.queue_time);
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->tsa.iostat.dev_busy_time %d\n",
+				       tsb->tsa.iostat.dev_busy_time);
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->tsa.iostat.dev_act_time %d\n",
+				       tsb->tsa.iostat.dev_act_time);
+			sense = tsb->tsa.iostat.sense;
+			break;
+		case 2: /* ts_ddpc */
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->tsa.ddpc.rc %d\n", tsb->tsa.ddpc.rc);
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+			       " tsb->tsa.ddpc.rcq:  ");
+			for (sl = 0; sl < 16; sl++) {
+				for (sct = 0; sct < 8; sct++) {
+					len += sprintf(page + len, " %02x",
+						       tsb->tsa.ddpc.rcq[sl]);
+				}
+				len += sprintf(page + len, "\n");
+			}
+			sense = tsb->tsa.ddpc.sense;
+			break;
+		case 3: /* tsa_intrg */
+			len += sprintf(page + len, KERN_ERR PRINTK_HEADER
+				      " tsb->tsa.intrg.: not supportet yet \n");
+			break;
+		}
+
+		if (sense) {
+			for (sl = 0; sl < 4; sl++) {
+				len += sprintf(page + len,
+					       KERN_ERR PRINTK_HEADER
+					       " Sense(hex) %2d-%2d:",
+					       (8 * sl), ((8 * sl) + 7));
+				for (sct = 0; sct < 8; sct++) {
+					len += sprintf(page + len, " %02x",
+						       sense[8 * sl + sct]);
+				}
+				len += sprintf(page + len, "\n");
+			}
+
+			if (sense[27] & DASD_SENSE_BIT_0) {
+				/* 24 Byte Sense Data */
+				sprintf(page + len, KERN_ERR PRINTK_HEADER
+					" 24 Byte: %x MSG %x, "
+					"%s MSGb to SYSOP\n",
+					sense[7] >> 4, sense[7] & 0x0f,
+					sense[1] & 0x10 ? "" : "no");
+			} else {
+				/* 32 Byte Sense Data */
+				sprintf(page + len, KERN_ERR PRINTK_HEADER
+					" 32 Byte: Format: %x "
+					"Exception class %x\n",
+					sense[6] & 0x0f, sense[22] >> 4);
+			}
+		} else {
+			sprintf(page + len, KERN_ERR PRINTK_HEADER
+				" SORRY - NO VALID SENSE AVAILABLE\n");
+		}
+	} else {
+		sprintf(page + len, KERN_ERR PRINTK_HEADER
+			" SORRY - NO TSB DATA AVAILABLE\n");
+	}
+	printk("%s", page);
+	free_page((unsigned long) page);
+}
+
+static void dasd_eckd_dump_sense(struct dasd_device *device,
+				 struct dasd_ccw_req *req, struct irb *irb)
+{
+	if (req && scsw_is_tm(&req->irb.scsw))
+		dasd_eckd_dump_sense_tcw(device, req, irb);
+	else
+		dasd_eckd_dump_sense_ccw(device, req, irb);
+}
+
+
 /*
  * max_blocks is dependent on the amount of storage that is available
  * in the static io buffer for each device. Currently each device has
diff --git a/drivers/s390/block/dasd_eckd.h b/drivers/s390/block/dasd_eckd.h
index eecfa776db15..ad45bcac3ce4 100644
--- a/drivers/s390/block/dasd_eckd.h
+++ b/drivers/s390/block/dasd_eckd.h
@@ -38,8 +38,11 @@
 #define DASD_ECKD_CCW_RELEASE		 0x94
 #define DASD_ECKD_CCW_READ_CKD_MT	 0x9e
 #define DASD_ECKD_CCW_WRITE_CKD_MT	 0x9d
+#define DASD_ECKD_CCW_WRITE_TRACK_DATA	 0xA5
+#define DASD_ECKD_CCW_READ_TRACK_DATA	 0xA6
 #define DASD_ECKD_CCW_RESERVE		 0xB4
 #define DASD_ECKD_CCW_PFX		 0xE7
+#define DASD_ECKD_CCW_PFX_READ		 0xEA
 #define DASD_ECKD_CCW_RSCK		 0xF9
 
 /*
@@ -123,7 +126,9 @@ struct DE_eckd_data {
 	unsigned long long ep_sys_time; /* Ext Parameter - System Time Stamp */
 	__u8 ep_format;        /* Extended Parameter format byte       */
 	__u8 ep_prio;          /* Extended Parameter priority I/O byte */
-	__u8 ep_reserved[6];   /* Extended Parameter Reserved          */
+	__u8 ep_reserved1;     /* Extended Parameter Reserved	       */
+	__u8 ep_rec_per_track; /* Number of records on a track	       */
+	__u8 ep_reserved[4];   /* Extended Parameter Reserved	       */
 } __attribute__ ((packed));
 
 struct LO_eckd_data {
@@ -144,11 +149,37 @@ struct LO_eckd_data {
 	__u16 length;
 } __attribute__ ((packed));
 
+struct LRE_eckd_data {
+	struct {
+		unsigned char orientation:2;
+		unsigned char operation:6;
+	} __attribute__ ((packed)) operation;
+	struct {
+		unsigned char length_valid:1;
+		unsigned char length_scope:1;
+		unsigned char imbedded_ccw_valid:1;
+		unsigned char check_bytes:2;
+		unsigned char imbedded_count_valid:1;
+		unsigned char reserved:1;
+		unsigned char read_count_suffix:1;
+	} __attribute__ ((packed)) auxiliary;
+	__u8 imbedded_ccw;
+	__u8 count;
+	struct ch_t seek_addr;
+	struct chr_t search_arg;
+	__u8 sector;
+	__u16 length;
+	__u8 imbedded_count;
+	__u8 extended_operation;
+	__u16 extended_parameter_length;
+	__u8 extended_parameter[0];
+} __attribute__ ((packed));
+
 /* Prefix data for format 0x00 and 0x01 */
 struct PFX_eckd_data {
 	unsigned char format;
 	struct {
-		unsigned char define_extend:1;
+		unsigned char define_extent:1;
 		unsigned char time_stamp:1;
 		unsigned char verify_base:1;
 		unsigned char hyper_pav:1;
@@ -158,9 +189,8 @@ struct PFX_eckd_data {
 	__u8 aux;
 	__u8 base_lss;
 	__u8 reserved[7];
-	struct DE_eckd_data define_extend;
-	struct LO_eckd_data locate_record;
-	__u8 LO_extended_data[4];
+	struct DE_eckd_data define_extent;
+	struct LRE_eckd_data locate_record;
 } __attribute__ ((packed));
 
 struct dasd_eckd_characteristics {
diff --git a/drivers/s390/block/dasd_eer.c b/drivers/s390/block/dasd_eer.c
index f8e05ce98621..9ce4209552ae 100644
--- a/drivers/s390/block/dasd_eer.c
+++ b/drivers/s390/block/dasd_eer.c
@@ -297,11 +297,12 @@ static void dasd_eer_write_standard_trigger(struct dasd_device *device,
 	struct dasd_eer_header header;
 	unsigned long flags;
 	struct eerbuffer *eerb;
+	char *sense;
 
 	/* go through cqr chain and count the valid sense data sets */
 	data_size = 0;
 	for (temp_cqr = cqr; temp_cqr; temp_cqr = temp_cqr->refers)
-		if (temp_cqr->irb.esw.esw0.erw.cons)
+		if (dasd_get_sense(&temp_cqr->irb))
 			data_size += 32;
 
 	header.total_size = sizeof(header) + data_size + 4; /* "EOR" */
@@ -316,9 +317,11 @@ static void dasd_eer_write_standard_trigger(struct dasd_device *device,
 	list_for_each_entry(eerb, &bufferlist, list) {
 		dasd_eer_start_record(eerb, header.total_size);
 		dasd_eer_write_buffer(eerb, (char *) &header, sizeof(header));
-		for (temp_cqr = cqr; temp_cqr; temp_cqr = temp_cqr->refers)
-			if (temp_cqr->irb.esw.esw0.erw.cons)
-				dasd_eer_write_buffer(eerb, cqr->irb.ecw, 32);
+		for (temp_cqr = cqr; temp_cqr; temp_cqr = temp_cqr->refers) {
+			sense = dasd_get_sense(&temp_cqr->irb);
+			if (sense)
+				dasd_eer_write_buffer(eerb, sense, 32);
+		}
 		dasd_eer_write_buffer(eerb, "EOR", 4);
 	}
 	spin_unlock_irqrestore(&bufferlock, flags);
@@ -451,6 +454,7 @@ int dasd_eer_enable(struct dasd_device *device)
 {
 	struct dasd_ccw_req *cqr;
 	unsigned long flags;
+	struct ccw1 *ccw;
 
 	if (device->eer_cqr)
 		return 0;
@@ -468,10 +472,11 @@ int dasd_eer_enable(struct dasd_device *device)
 	cqr->expires = 10 * HZ;
 	clear_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
 
-	cqr->cpaddr->cmd_code = DASD_ECKD_CCW_SNSS;
-	cqr->cpaddr->count = SNSS_DATA_SIZE;
-	cqr->cpaddr->flags = 0;
-	cqr->cpaddr->cda = (__u32)(addr_t) cqr->data;
+	ccw = cqr->cpaddr;
+	ccw->cmd_code = DASD_ECKD_CCW_SNSS;
+	ccw->count = SNSS_DATA_SIZE;
+	ccw->flags = 0;
+	ccw->cda = (__u32)(addr_t) cqr->data;
 
 	cqr->buildclk = get_clock();
 	cqr->status = DASD_CQR_FILLED;
diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h
index 29991a9fa07a..7b314c1d471e 100644
--- a/drivers/s390/block/dasd_int.h
+++ b/drivers/s390/block/dasd_int.h
@@ -157,7 +157,8 @@ struct dasd_ccw_req {
 	struct dasd_block *block;	/* the originating block device */
 	struct dasd_device *memdev;	/* the device used to allocate this */
 	struct dasd_device *startdev;	/* device the request is started on */
-	struct ccw1 *cpaddr;		/* address of channel program */
+	void *cpaddr;			/* address of ccw or tcw */
+	unsigned char cpmode;		/* 0 = cmd mode, 1 = itcw */
 	char status;			/* status of this request */
 	short retries;			/* A retry counter */
 	unsigned long flags;        	/* flags of this request */
@@ -573,12 +574,14 @@ int dasd_generic_notify(struct ccw_device *, int);
 void dasd_generic_handle_state_change(struct dasd_device *);
 
 int dasd_generic_read_dev_chars(struct dasd_device *, char *, void **, int);
+char *dasd_get_sense(struct irb *);
 
 /* externals in dasd_devmap.c */
 extern int dasd_max_devindex;
 extern int dasd_probeonly;
 extern int dasd_autodetect;
 extern int dasd_nopav;
+extern int dasd_nofcx;
 
 int dasd_devmap_init(void);
 void dasd_devmap_exit(void);
-- 
cgit v1.2.3-55-g7522


From 94f5b09d97ee1f803c76d0262e0b0d3791825d09 Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:23:50 +0100
Subject: [S390] move sysinfo.c from drivers/s390 to arch/s390/kernel

All in sysinfo.c is core kernel code and not driver code. So move it
to arch/s390/kernel. Also includes some small cleanups.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/Makefile  |   4 +-
 arch/s390/kernel/sysinfo.c | 428 +++++++++++++++++++++++++++++++++++++++++
 drivers/s390/Makefile      |   4 +-
 drivers/s390/sysinfo.c     | 469 ---------------------------------------------
 4 files changed, 432 insertions(+), 473 deletions(-)
 create mode 100644 arch/s390/kernel/sysinfo.c
 delete mode 100644 drivers/s390/sysinfo.c

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index 3edc6c6f258b..33e7aee70513 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -17,10 +17,12 @@ CFLAGS_smp.o	:= -Wno-nonnull
 #
 CFLAGS_ptrace.o		+= -DUTS_MACHINE='"$(UTS_MACHINE)"'
 
+CFLAGS_sysinfo.o += -Iinclude/math-emu -Iarch/s390/math-emu -w
+
 obj-y	:=  bitmap.o traps.o time.o process.o base.o early.o setup.o \
 	    processor.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \
 	    s390_ext.o debug.o irq.o ipl.o dis.o diag.o mem_detect.o \
-	    vdso.o vtime.o
+	    vdso.o vtime.o sysinfo.o
 
 obj-y	+= $(if $(CONFIG_64BIT),entry64.o,entry.o)
 obj-y	+= $(if $(CONFIG_64BIT),reipl64.o,reipl.o)
diff --git a/arch/s390/kernel/sysinfo.c b/arch/s390/kernel/sysinfo.c
new file mode 100644
index 000000000000..b5e75e1061c8
--- /dev/null
+++ b/arch/s390/kernel/sysinfo.c
@@ -0,0 +1,428 @@
+/*
+ *  Copyright IBM Corp. 2001, 2009
+ *  Author(s): Ulrich Weigand <Ulrich.Weigand@de.ibm.com>,
+ *	       Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ */
+
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <asm/ebcdic.h>
+#include <asm/sysinfo.h>
+#include <asm/cpcmd.h>
+
+/* Sigh, math-emu. Don't ask. */
+#include <asm/sfp-util.h>
+#include <math-emu/soft-fp.h>
+#include <math-emu/single.h>
+
+static inline int stsi_0(void)
+{
+	int rc = stsi(NULL, 0, 0, 0);
+	return rc == -ENOSYS ? rc : (((unsigned int) rc) >> 28);
+}
+
+static int stsi_1_1_1(struct sysinfo_1_1_1 *info, char *page, int len)
+{
+	if (stsi(info, 1, 1, 1) == -ENOSYS)
+		return len;
+
+	EBCASC(info->manufacturer, sizeof(info->manufacturer));
+	EBCASC(info->type, sizeof(info->type));
+	EBCASC(info->model, sizeof(info->model));
+	EBCASC(info->sequence, sizeof(info->sequence));
+	EBCASC(info->plant, sizeof(info->plant));
+	EBCASC(info->model_capacity, sizeof(info->model_capacity));
+	EBCASC(info->model_perm_cap, sizeof(info->model_perm_cap));
+	EBCASC(info->model_temp_cap, sizeof(info->model_temp_cap));
+	len += sprintf(page + len, "Manufacturer:         %-16.16s\n",
+		       info->manufacturer);
+	len += sprintf(page + len, "Type:                 %-4.4s\n",
+		       info->type);
+	if (info->model[0] != '\0')
+		/*
+		 * Sigh: the model field has been renamed with System z9
+		 * to model_capacity and a new model field has been added
+		 * after the plant field. To avoid confusing older programs
+		 * the "Model:" prints "model_capacity model" or just
+		 * "model_capacity" if the model string is empty .
+		 */
+		len += sprintf(page + len,
+			       "Model:                %-16.16s %-16.16s\n",
+			       info->model_capacity, info->model);
+	else
+		len += sprintf(page + len, "Model:                %-16.16s\n",
+			       info->model_capacity);
+	len += sprintf(page + len, "Sequence Code:        %-16.16s\n",
+		       info->sequence);
+	len += sprintf(page + len, "Plant:                %-4.4s\n",
+		       info->plant);
+	len += sprintf(page + len, "Model Capacity:       %-16.16s %08u\n",
+		       info->model_capacity, *(u32 *) info->model_cap_rating);
+	if (info->model_perm_cap[0] != '\0')
+		len += sprintf(page + len,
+			       "Model Perm. Capacity: %-16.16s %08u\n",
+			       info->model_perm_cap,
+			       *(u32 *) info->model_perm_cap_rating);
+	if (info->model_temp_cap[0] != '\0')
+		len += sprintf(page + len,
+			       "Model Temp. Capacity: %-16.16s %08u\n",
+			       info->model_temp_cap,
+			       *(u32 *) info->model_temp_cap_rating);
+	return len;
+}
+
+static int stsi_1_2_2(struct sysinfo_1_2_2 *info, char *page, int len)
+{
+	struct sysinfo_1_2_2_extension *ext;
+	int i;
+
+	if (stsi(info, 1, 2, 2) == -ENOSYS)
+		return len;
+	ext = (struct sysinfo_1_2_2_extension *)
+		((unsigned long) info + info->acc_offset);
+
+	len += sprintf(page + len, "\n");
+	len += sprintf(page + len, "CPUs Total:           %d\n",
+		       info->cpus_total);
+	len += sprintf(page + len, "CPUs Configured:      %d\n",
+		       info->cpus_configured);
+	len += sprintf(page + len, "CPUs Standby:         %d\n",
+		       info->cpus_standby);
+	len += sprintf(page + len, "CPUs Reserved:        %d\n",
+		       info->cpus_reserved);
+
+	if (info->format == 1) {
+		/*
+		 * Sigh 2. According to the specification the alternate
+		 * capability field is a 32 bit floating point number
+		 * if the higher order 8 bits are not zero. Printing
+		 * a floating point number in the kernel is a no-no,
+		 * always print the number as 32 bit unsigned integer.
+		 * The user-space needs to know about the strange
+		 * encoding of the alternate cpu capability.
+		 */
+		len += sprintf(page + len, "Capability:           %u %u\n",
+			       info->capability, ext->alt_capability);
+		for (i = 2; i <= info->cpus_total; i++)
+			len += sprintf(page + len,
+				       "Adjustment %02d-way:    %u %u\n",
+				       i, info->adjustment[i-2],
+				       ext->alt_adjustment[i-2]);
+
+	} else {
+		len += sprintf(page + len, "Capability:           %u\n",
+			       info->capability);
+		for (i = 2; i <= info->cpus_total; i++)
+			len += sprintf(page + len,
+				       "Adjustment %02d-way:    %u\n",
+				       i, info->adjustment[i-2]);
+	}
+
+	if (info->secondary_capability != 0)
+		len += sprintf(page + len, "Secondary Capability: %d\n",
+			       info->secondary_capability);
+	return len;
+}
+
+static int stsi_2_2_2(struct sysinfo_2_2_2 *info, char *page, int len)
+{
+	if (stsi(info, 2, 2, 2) == -ENOSYS)
+		return len;
+
+	EBCASC(info->name, sizeof(info->name));
+
+	len += sprintf(page + len, "\n");
+	len += sprintf(page + len, "LPAR Number:          %d\n",
+		       info->lpar_number);
+
+	len += sprintf(page + len, "LPAR Characteristics: ");
+	if (info->characteristics & LPAR_CHAR_DEDICATED)
+		len += sprintf(page + len, "Dedicated ");
+	if (info->characteristics & LPAR_CHAR_SHARED)
+		len += sprintf(page + len, "Shared ");
+	if (info->characteristics & LPAR_CHAR_LIMITED)
+		len += sprintf(page + len, "Limited ");
+	len += sprintf(page + len, "\n");
+
+	len += sprintf(page + len, "LPAR Name:            %-8.8s\n",
+		       info->name);
+
+	len += sprintf(page + len, "LPAR Adjustment:      %d\n",
+		       info->caf);
+
+	len += sprintf(page + len, "LPAR CPUs Total:      %d\n",
+		       info->cpus_total);
+	len += sprintf(page + len, "LPAR CPUs Configured: %d\n",
+		       info->cpus_configured);
+	len += sprintf(page + len, "LPAR CPUs Standby:    %d\n",
+		       info->cpus_standby);
+	len += sprintf(page + len, "LPAR CPUs Reserved:   %d\n",
+		       info->cpus_reserved);
+	len += sprintf(page + len, "LPAR CPUs Dedicated:  %d\n",
+		       info->cpus_dedicated);
+	len += sprintf(page + len, "LPAR CPUs Shared:     %d\n",
+		       info->cpus_shared);
+	return len;
+}
+
+static int stsi_3_2_2(struct sysinfo_3_2_2 *info, char *page, int len)
+{
+	int i;
+
+	if (stsi(info, 3, 2, 2) == -ENOSYS)
+		return len;
+	for (i = 0; i < info->count; i++) {
+		EBCASC(info->vm[i].name, sizeof(info->vm[i].name));
+		EBCASC(info->vm[i].cpi, sizeof(info->vm[i].cpi));
+		len += sprintf(page + len, "\n");
+		len += sprintf(page + len, "VM%02d Name:            %-8.8s\n",
+			       i, info->vm[i].name);
+		len += sprintf(page + len, "VM%02d Control Program: %-16.16s\n",
+			       i, info->vm[i].cpi);
+
+		len += sprintf(page + len, "VM%02d Adjustment:      %d\n",
+			       i, info->vm[i].caf);
+
+		len += sprintf(page + len, "VM%02d CPUs Total:      %d\n",
+			       i, info->vm[i].cpus_total);
+		len += sprintf(page + len, "VM%02d CPUs Configured: %d\n",
+			       i, info->vm[i].cpus_configured);
+		len += sprintf(page + len, "VM%02d CPUs Standby:    %d\n",
+			       i, info->vm[i].cpus_standby);
+		len += sprintf(page + len, "VM%02d CPUs Reserved:   %d\n",
+			       i, info->vm[i].cpus_reserved);
+	}
+	return len;
+}
+
+static int proc_read_sysinfo(char *page, char **start,
+			     off_t off, int count,
+			     int *eof, void *data)
+{
+	unsigned long info = get_zeroed_page(GFP_KERNEL);
+	int level, len;
+
+	if (!info)
+		return 0;
+
+	len = 0;
+	level = stsi_0();
+	if (level >= 1)
+		len = stsi_1_1_1((struct sysinfo_1_1_1 *) info, page, len);
+
+	if (level >= 1)
+		len = stsi_1_2_2((struct sysinfo_1_2_2 *) info, page, len);
+
+	if (level >= 2)
+		len = stsi_2_2_2((struct sysinfo_2_2_2 *) info, page, len);
+
+	if (level >= 3)
+		len = stsi_3_2_2((struct sysinfo_3_2_2 *) info, page, len);
+
+	free_page(info);
+	return len;
+}
+
+static __init int create_proc_sysinfo(void)
+{
+	create_proc_read_entry("sysinfo", 0444, NULL,
+			       proc_read_sysinfo, NULL);
+	return 0;
+}
+device_initcall(create_proc_sysinfo);
+
+/*
+ * Service levels interface.
+ */
+
+static DECLARE_RWSEM(service_level_sem);
+static LIST_HEAD(service_level_list);
+
+int register_service_level(struct service_level *slr)
+{
+	struct service_level *ptr;
+
+	down_write(&service_level_sem);
+	list_for_each_entry(ptr, &service_level_list, list)
+		if (ptr == slr) {
+			up_write(&service_level_sem);
+			return -EEXIST;
+		}
+	list_add_tail(&slr->list, &service_level_list);
+	up_write(&service_level_sem);
+	return 0;
+}
+EXPORT_SYMBOL(register_service_level);
+
+int unregister_service_level(struct service_level *slr)
+{
+	struct service_level *ptr, *next;
+	int rc = -ENOENT;
+
+	down_write(&service_level_sem);
+	list_for_each_entry_safe(ptr, next, &service_level_list, list) {
+		if (ptr != slr)
+			continue;
+		list_del(&ptr->list);
+		rc = 0;
+		break;
+	}
+	up_write(&service_level_sem);
+	return rc;
+}
+EXPORT_SYMBOL(unregister_service_level);
+
+static void *service_level_start(struct seq_file *m, loff_t *pos)
+{
+	down_read(&service_level_sem);
+	return seq_list_start(&service_level_list, *pos);
+}
+
+static void *service_level_next(struct seq_file *m, void *p, loff_t *pos)
+{
+	return seq_list_next(p, &service_level_list, pos);
+}
+
+static void service_level_stop(struct seq_file *m, void *p)
+{
+	up_read(&service_level_sem);
+}
+
+static int service_level_show(struct seq_file *m, void *p)
+{
+	struct service_level *slr;
+
+	slr = list_entry(p, struct service_level, list);
+	slr->seq_print(m, slr);
+	return 0;
+}
+
+static const struct seq_operations service_level_seq_ops = {
+	.start		= service_level_start,
+	.next		= service_level_next,
+	.stop		= service_level_stop,
+	.show		= service_level_show
+};
+
+static int service_level_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &service_level_seq_ops);
+}
+
+static const struct file_operations service_level_ops = {
+	.open		= service_level_open,
+	.read		= seq_read,
+	.llseek 	= seq_lseek,
+	.release	= seq_release
+};
+
+static void service_level_vm_print(struct seq_file *m,
+				   struct service_level *slr)
+{
+	char *query_buffer, *str;
+
+	query_buffer = kmalloc(1024, GFP_KERNEL | GFP_DMA);
+	if (!query_buffer)
+		return;
+	cpcmd("QUERY CPLEVEL", query_buffer, 1024, NULL);
+	str = strchr(query_buffer, '\n');
+	if (str)
+		*str = 0;
+	seq_printf(m, "VM: %s\n", query_buffer);
+	kfree(query_buffer);
+}
+
+static struct service_level service_level_vm = {
+	.seq_print = service_level_vm_print
+};
+
+static __init int create_proc_service_level(void)
+{
+	proc_create("service_levels", 0, NULL, &service_level_ops);
+	if (MACHINE_IS_VM)
+		register_service_level(&service_level_vm);
+	return 0;
+}
+subsys_initcall(create_proc_service_level);
+
+/*
+ * Bogomips calculation based on cpu capability.
+ */
+int get_cpu_capability(unsigned int *capability)
+{
+	struct sysinfo_1_2_2 *info;
+	int rc;
+
+	info = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!info)
+		return -ENOMEM;
+	rc = stsi(info, 1, 2, 2);
+	if (rc == -ENOSYS)
+		goto out;
+	rc = 0;
+	*capability = info->capability;
+out:
+	free_page((unsigned long) info);
+	return rc;
+}
+
+/*
+ * CPU capability might have changed. Therefore recalculate loops_per_jiffy.
+ */
+void s390_adjust_jiffies(void)
+{
+	struct sysinfo_1_2_2 *info;
+	const unsigned int fmil = 0x4b189680;	/* 1e7 as 32-bit float. */
+	FP_DECL_S(SA); FP_DECL_S(SB); FP_DECL_S(SR);
+	FP_DECL_EX;
+	unsigned int capability;
+
+	info = (void *) get_zeroed_page(GFP_KERNEL);
+	if (!info)
+		return;
+
+	if (stsi(info, 1, 2, 2) != -ENOSYS) {
+		/*
+		 * Major sigh. The cpu capability encoding is "special".
+		 * If the first 9 bits of info->capability are 0 then it
+		 * is a 32 bit unsigned integer in the range 0 .. 2^23.
+		 * If the first 9 bits are != 0 then it is a 32 bit float.
+		 * In addition a lower value indicates a proportionally
+		 * higher cpu capacity. Bogomips are the other way round.
+		 * To get to a halfway suitable number we divide 1e7
+		 * by the cpu capability number. Yes, that means a floating
+		 * point division .. math-emu here we come :-)
+		 */
+		FP_UNPACK_SP(SA, &fmil);
+		if ((info->capability >> 23) == 0)
+			FP_FROM_INT_S(SB, info->capability, 32, int);
+		else
+			FP_UNPACK_SP(SB, &info->capability);
+		FP_DIV_S(SR, SA, SB);
+		FP_TO_INT_S(capability, SR, 32, 0);
+	} else
+		/*
+		 * Really old machine without stsi block for basic
+		 * cpu information. Report 42.0 bogomips.
+		 */
+		capability = 42;
+	loops_per_jiffy = capability * (500000/HZ);
+	free_page((unsigned long) info);
+}
+
+/*
+ * calibrate the delay loop
+ */
+void __cpuinit calibrate_delay(void)
+{
+	s390_adjust_jiffies();
+	/* Print the good old Bogomips line .. */
+	printk(KERN_DEBUG "Calibrating delay loop (skipped)... "
+	       "%lu.%02lu BogoMIPS preset\n", loops_per_jiffy/(500000/HZ),
+	       (loops_per_jiffy/(5000/HZ)) % 100);
+}
diff --git a/drivers/s390/Makefile b/drivers/s390/Makefile
index d0eae59bc366..0828dc839355 100644
--- a/drivers/s390/Makefile
+++ b/drivers/s390/Makefile
@@ -2,9 +2,7 @@
 # Makefile for the S/390 specific device drivers
 #
 
-CFLAGS_sysinfo.o += -Iinclude/math-emu -Iarch/s390/math-emu -w
-
-obj-y += s390mach.o sysinfo.o
+obj-y += s390mach.o
 obj-y += cio/ block/ char/ crypto/ net/ scsi/ kvm/
 
 drivers-y += drivers/s390/built-in.o
diff --git a/drivers/s390/sysinfo.c b/drivers/s390/sysinfo.c
deleted file mode 100644
index 0eea90781385..000000000000
--- a/drivers/s390/sysinfo.c
+++ /dev/null
@@ -1,469 +0,0 @@
-/*
- *  drivers/s390/sysinfo.c
- *
- *  Copyright IBM Corp. 2001, 2008
- *  Author(s): Ulrich Weigand (Ulrich.Weigand@de.ibm.com)
- *	       Martin Schwidefsky <schwidefsky@de.ibm.com>
- */
-
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/proc_fs.h>
-#include <linux/seq_file.h>
-#include <linux/init.h>
-#include <linux/delay.h>
-#include <linux/module.h>
-#include <asm/ebcdic.h>
-#include <asm/sysinfo.h>
-#include <asm/cpcmd.h>
-
-/* Sigh, math-emu. Don't ask. */
-#include <asm/sfp-util.h>
-#include <math-emu/soft-fp.h>
-#include <math-emu/single.h>
-
-static inline int stsi_0(void)
-{
-	int rc = stsi (NULL, 0, 0, 0);
-	return rc == -ENOSYS ? rc : (((unsigned int) rc) >> 28);
-}
-
-static int stsi_1_1_1(struct sysinfo_1_1_1 *info, char *page, int len)
-{
-	if (stsi(info, 1, 1, 1) == -ENOSYS)
-		return len;
-
-	EBCASC(info->manufacturer, sizeof(info->manufacturer));
-	EBCASC(info->type, sizeof(info->type));
-	EBCASC(info->model, sizeof(info->model));
-	EBCASC(info->sequence, sizeof(info->sequence));
-	EBCASC(info->plant, sizeof(info->plant));
-	EBCASC(info->model_capacity, sizeof(info->model_capacity));
-	EBCASC(info->model_perm_cap, sizeof(info->model_perm_cap));
-	EBCASC(info->model_temp_cap, sizeof(info->model_temp_cap));
-	len += sprintf(page + len, "Manufacturer:         %-16.16s\n",
-		       info->manufacturer);
-	len += sprintf(page + len, "Type:                 %-4.4s\n",
-		       info->type);
-	if (info->model[0] != '\0')
-		/*
-		 * Sigh: the model field has been renamed with System z9
-		 * to model_capacity and a new model field has been added
-		 * after the plant field. To avoid confusing older programs
-		 * the "Model:" prints "model_capacity model" or just
-		 * "model_capacity" if the model string is empty .
-		 */
-		len += sprintf(page + len,
-			       "Model:                %-16.16s %-16.16s\n",
-			       info->model_capacity, info->model);
-	else
-		len += sprintf(page + len, "Model:                %-16.16s\n",
-			       info->model_capacity);
-	len += sprintf(page + len, "Sequence Code:        %-16.16s\n",
-		       info->sequence);
-	len += sprintf(page + len, "Plant:                %-4.4s\n",
-		       info->plant);
-	len += sprintf(page + len, "Model Capacity:       %-16.16s %08u\n",
-		       info->model_capacity, *(u32 *) info->model_cap_rating);
-	if (info->model_perm_cap[0] != '\0')
-		len += sprintf(page + len,
-			       "Model Perm. Capacity: %-16.16s %08u\n",
-			       info->model_perm_cap,
-			       *(u32 *) info->model_perm_cap_rating);
-	if (info->model_temp_cap[0] != '\0')
-		len += sprintf(page + len,
-			       "Model Temp. Capacity: %-16.16s %08u\n",
-			       info->model_temp_cap,
-			       *(u32 *) info->model_temp_cap_rating);
-	return len;
-}
-
-#if 0 /* Currently unused */
-static int stsi_1_2_1(struct sysinfo_1_2_1 *info, char *page, int len)
-{
-	if (stsi(info, 1, 2, 1) == -ENOSYS)
-		return len;
-
-	len += sprintf(page + len, "\n");
-	EBCASC(info->sequence, sizeof(info->sequence));
-	EBCASC(info->plant, sizeof(info->plant));
-	len += sprintf(page + len, "Sequence Code of CPU: %-16.16s\n",
-		       info->sequence);
-	len += sprintf(page + len, "Plant of CPU:         %-16.16s\n",
-		       info->plant);
-	return len;
-}
-#endif
-
-static int stsi_1_2_2(struct sysinfo_1_2_2 *info, char *page, int len)
-{
-	struct sysinfo_1_2_2_extension *ext;
-	int i;
-
-	if (stsi(info, 1, 2, 2) == -ENOSYS)
-		return len;
-	ext = (struct sysinfo_1_2_2_extension *)
-		((unsigned long) info + info->acc_offset);
-
-	len += sprintf(page + len, "\n");
-	len += sprintf(page + len, "CPUs Total:           %d\n",
-		       info->cpus_total);
-	len += sprintf(page + len, "CPUs Configured:      %d\n",
-		       info->cpus_configured);
-	len += sprintf(page + len, "CPUs Standby:         %d\n",
-		       info->cpus_standby);
-	len += sprintf(page + len, "CPUs Reserved:        %d\n",
-		       info->cpus_reserved);
-
-	if (info->format == 1) {
-		/*
-		 * Sigh 2. According to the specification the alternate
-		 * capability field is a 32 bit floating point number
-		 * if the higher order 8 bits are not zero. Printing
-		 * a floating point number in the kernel is a no-no,
-		 * always print the number as 32 bit unsigned integer.
-		 * The user-space needs to know about the strange
-		 * encoding of the alternate cpu capability.
-		 */
-		len += sprintf(page + len, "Capability:           %u %u\n",
-			       info->capability, ext->alt_capability);
-		for (i = 2; i <= info->cpus_total; i++)
-			len += sprintf(page + len,
-				       "Adjustment %02d-way:    %u %u\n",
-				       i, info->adjustment[i-2],
-				       ext->alt_adjustment[i-2]);
-
-	} else {
-		len += sprintf(page + len, "Capability:           %u\n",
-			       info->capability);
-		for (i = 2; i <= info->cpus_total; i++)
-			len += sprintf(page + len,
-				       "Adjustment %02d-way:    %u\n",
-				       i, info->adjustment[i-2]);
-	}
-
-	if (info->secondary_capability != 0)
-		len += sprintf(page + len, "Secondary Capability: %d\n",
-			       info->secondary_capability);
-
-	return len;
-}
-
-#if 0 /* Currently unused */
-static int stsi_2_2_1(struct sysinfo_2_2_1 *info, char *page, int len)
-{
-	if (stsi(info, 2, 2, 1) == -ENOSYS)
-		return len;
-
-	len += sprintf(page + len, "\n");
-	EBCASC (info->sequence, sizeof(info->sequence));
-	EBCASC (info->plant, sizeof(info->plant));
-	len += sprintf(page + len, "Sequence Code of logical CPU: %-16.16s\n",
-		       info->sequence);
-	len += sprintf(page + len, "Plant of logical CPU: %-16.16s\n",
-		       info->plant);
-	return len;
-}
-#endif
-
-static int stsi_2_2_2(struct sysinfo_2_2_2 *info, char *page, int len)
-{
-	if (stsi(info, 2, 2, 2) == -ENOSYS)
-		return len;
-
-	EBCASC (info->name, sizeof(info->name));
-
-	len += sprintf(page + len, "\n");
-	len += sprintf(page + len, "LPAR Number:          %d\n",
-		       info->lpar_number);
-
-	len += sprintf(page + len, "LPAR Characteristics: ");
-	if (info->characteristics & LPAR_CHAR_DEDICATED)
-		len += sprintf(page + len, "Dedicated ");
-	if (info->characteristics & LPAR_CHAR_SHARED)
-		len += sprintf(page + len, "Shared ");
-	if (info->characteristics & LPAR_CHAR_LIMITED)
-		len += sprintf(page + len, "Limited ");
-	len += sprintf(page + len, "\n");
-
-	len += sprintf(page + len, "LPAR Name:            %-8.8s\n",
-		       info->name);
-
-	len += sprintf(page + len, "LPAR Adjustment:      %d\n",
-		       info->caf);
-
-	len += sprintf(page + len, "LPAR CPUs Total:      %d\n",
-		       info->cpus_total);
-	len += sprintf(page + len, "LPAR CPUs Configured: %d\n",
-		       info->cpus_configured);
-	len += sprintf(page + len, "LPAR CPUs Standby:    %d\n",
-		       info->cpus_standby);
-	len += sprintf(page + len, "LPAR CPUs Reserved:   %d\n",
-		       info->cpus_reserved);
-	len += sprintf(page + len, "LPAR CPUs Dedicated:  %d\n",
-		       info->cpus_dedicated);
-	len += sprintf(page + len, "LPAR CPUs Shared:     %d\n",
-		       info->cpus_shared);
-	return len;
-}
-
-static int stsi_3_2_2(struct sysinfo_3_2_2 *info, char *page, int len)
-{
-	int i;
-
-	if (stsi(info, 3, 2, 2) == -ENOSYS)
-		return len;
-	for (i = 0; i < info->count; i++) {
-		EBCASC (info->vm[i].name, sizeof(info->vm[i].name));
-		EBCASC (info->vm[i].cpi, sizeof(info->vm[i].cpi));
-		len += sprintf(page + len, "\n");
-		len += sprintf(page + len, "VM%02d Name:            %-8.8s\n",
-			       i, info->vm[i].name);
-		len += sprintf(page + len, "VM%02d Control Program: %-16.16s\n",
-			       i, info->vm[i].cpi);
-
-		len += sprintf(page + len, "VM%02d Adjustment:      %d\n",
-			       i, info->vm[i].caf);
-
-		len += sprintf(page + len, "VM%02d CPUs Total:      %d\n",
-			       i, info->vm[i].cpus_total);
-		len += sprintf(page + len, "VM%02d CPUs Configured: %d\n",
-			       i, info->vm[i].cpus_configured);
-		len += sprintf(page + len, "VM%02d CPUs Standby:    %d\n",
-			       i, info->vm[i].cpus_standby);
-		len += sprintf(page + len, "VM%02d CPUs Reserved:   %d\n",
-			       i, info->vm[i].cpus_reserved);
-	}
-	return len;
-}
-
-
-static int proc_read_sysinfo(char *page, char **start,
-                             off_t off, int count,
-                             int *eof, void *data)
-{
-	unsigned long info = get_zeroed_page (GFP_KERNEL);
-	int level, len;
-	
-	if (!info)
-		return 0;
-
-	len = 0;
-	level = stsi_0();
-	if (level >= 1)
-		len = stsi_1_1_1((struct sysinfo_1_1_1 *) info, page, len);
-
-	if (level >= 1)
-		len = stsi_1_2_2((struct sysinfo_1_2_2 *) info, page, len);
-
-	if (level >= 2)
-		len = stsi_2_2_2((struct sysinfo_2_2_2 *) info, page, len);
-
-	if (level >= 3)
-		len = stsi_3_2_2((struct sysinfo_3_2_2 *) info, page, len);
-
-	free_page (info);
-        return len;
-}
-
-static __init int create_proc_sysinfo(void)
-{
-	create_proc_read_entry("sysinfo", 0444, NULL,
-			       proc_read_sysinfo, NULL);
-	return 0;
-}
-
-__initcall(create_proc_sysinfo);
-
-/*
- * Service levels interface.
- */
-
-static DECLARE_RWSEM(service_level_sem);
-static LIST_HEAD(service_level_list);
-
-int register_service_level(struct service_level *slr)
-{
-	struct service_level *ptr;
-
-	down_write(&service_level_sem);
-	list_for_each_entry(ptr, &service_level_list, list)
-		if (ptr == slr) {
-			up_write(&service_level_sem);
-			return -EEXIST;
-		}
-	list_add_tail(&slr->list, &service_level_list);
-	up_write(&service_level_sem);
-	return 0;
-}
-EXPORT_SYMBOL(register_service_level);
-
-int unregister_service_level(struct service_level *slr)
-{
-	struct service_level *ptr, *next;
-	int rc = -ENOENT;
-
-	down_write(&service_level_sem);
-	list_for_each_entry_safe(ptr, next, &service_level_list, list) {
-		if (ptr != slr)
-			continue;
-		list_del(&ptr->list);
-		rc = 0;
-		break;
-	}
-	up_write(&service_level_sem);
-	return rc;
-}
-EXPORT_SYMBOL(unregister_service_level);
-
-static void *service_level_start(struct seq_file *m, loff_t *pos)
-{
-	down_read(&service_level_sem);
-	return seq_list_start(&service_level_list, *pos);
-}
-
-static void *service_level_next(struct seq_file *m, void *p, loff_t *pos)
-{
-	return seq_list_next(p, &service_level_list, pos);
-}
-
-static void service_level_stop(struct seq_file *m, void *p)
-{
-	up_read(&service_level_sem);
-}
-
-static int service_level_show(struct seq_file *m, void *p)
-{
-	struct service_level *slr;
-
-	slr = list_entry(p, struct service_level, list);
-	slr->seq_print(m, slr);
-	return 0;
-}
-
-static const struct seq_operations service_level_seq_ops = {
-	.start		= service_level_start,
-	.next		= service_level_next,
-	.stop		= service_level_stop,
-	.show		= service_level_show
-};
-
-static int service_level_open(struct inode *inode, struct file *file)
-{
-	return seq_open(file, &service_level_seq_ops);
-}
-
-static const struct file_operations service_level_ops = {
-	.open		= service_level_open,
-	.read		= seq_read,
-	.llseek 	= seq_lseek,
-	.release	= seq_release
-};
-
-static void service_level_vm_print(struct seq_file *m,
-				   struct service_level *slr)
-{
-	char *query_buffer, *str;
-
-	query_buffer = kmalloc(1024, GFP_KERNEL | GFP_DMA);
-	if (!query_buffer)
-		return;
-	cpcmd("QUERY CPLEVEL", query_buffer, 1024, NULL);
-	str = strchr(query_buffer, '\n');
-	if (str)
-		*str = 0;
-	seq_printf(m, "VM: %s\n", query_buffer);
-	kfree(query_buffer);
-}
-
-static struct service_level service_level_vm = {
-	.seq_print = service_level_vm_print
-};
-
-static __init int create_proc_service_level(void)
-{
-	proc_create("service_levels", 0, NULL, &service_level_ops);
-	if (MACHINE_IS_VM)
-		register_service_level(&service_level_vm);
-	return 0;
-}
-
-subsys_initcall(create_proc_service_level);
-
-/*
- * Bogomips calculation based on cpu capability.
- */
-
-int get_cpu_capability(unsigned int *capability)
-{
-	struct sysinfo_1_2_2 *info;
-	int rc;
-
-	info = (void *) get_zeroed_page(GFP_KERNEL);
-	if (!info)
-		return -ENOMEM;
-	rc = stsi(info, 1, 2, 2);
-	if (rc == -ENOSYS)
-		goto out;
-	rc = 0;
-	*capability = info->capability;
-out:
-	free_page((unsigned long) info);
-	return rc;
-}
-
-/*
- * CPU capability might have changed. Therefore recalculate loops_per_jiffy.
- */
-void s390_adjust_jiffies(void)
-{
-	struct sysinfo_1_2_2 *info;
-	const unsigned int fmil = 0x4b189680;	/* 1e7 as 32-bit float. */
-	FP_DECL_S(SA); FP_DECL_S(SB); FP_DECL_S(SR);
-	FP_DECL_EX;
-	unsigned int capability;
-
-	info = (void *) get_zeroed_page(GFP_KERNEL);
-	if (!info)
-		return;
-
-	if (stsi(info, 1, 2, 2) != -ENOSYS) {
-		/*
-		 * Major sigh. The cpu capability encoding is "special".
-		 * If the first 9 bits of info->capability are 0 then it
-		 * is a 32 bit unsigned integer in the range 0 .. 2^23.
-		 * If the first 9 bits are != 0 then it is a 32 bit float.
-		 * In addition a lower value indicates a proportionally
-		 * higher cpu capacity. Bogomips are the other way round.
-		 * To get to a halfway suitable number we divide 1e7
-		 * by the cpu capability number. Yes, that means a floating
-		 * point division .. math-emu here we come :-)
-		 */
-		FP_UNPACK_SP(SA, &fmil);
-		if ((info->capability >> 23) == 0)
-			FP_FROM_INT_S(SB, info->capability, 32, int);
-		else
-			FP_UNPACK_SP(SB, &info->capability);
-		FP_DIV_S(SR, SA, SB);
-		FP_TO_INT_S(capability, SR, 32, 0);
-	} else
-		/*
-		 * Really old machine without stsi block for basic
-		 * cpu information. Report 42.0 bogomips.
-		 */
-		capability = 42;
-	loops_per_jiffy = capability * (500000/HZ);
-	free_page((unsigned long) info);
-}
-
-/*
- * calibrate the delay loop
- */
-void __cpuinit calibrate_delay(void)
-{
-	s390_adjust_jiffies();
-	/* Print the good old Bogomips line .. */
-	printk(KERN_DEBUG "Calibrating delay loop (skipped)... "
-	       "%lu.%02lu BogoMIPS preset\n", loops_per_jiffy/(500000/HZ),
-	       (loops_per_jiffy/(5000/HZ)) % 100);
-}
-- 
cgit v1.2.3-55-g7522


From cbdc229245e8cf5c201e68221ebf2f33d2aaf029 Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:23:52 +0100
Subject: [S390] arch/s390/kernel/process.c: fix whitespace damage

Fix all the whitespace damage in process.c, especially copy_thread().
Next patch will add code to copy_thread() which needs to 'fixed' first.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/process.c | 64 +++++++++++++++++++---------------------------
 1 file changed, 27 insertions(+), 37 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 5cd38a90e64d..e0563baa1cfe 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -1,18 +1,10 @@
 /*
- *  arch/s390/kernel/process.c
+ * This file handles the architecture dependent parts of process handling.
  *
- *  S390 version
- *    Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
- *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
- *               Hartmut Penner (hp@de.ibm.com),
- *               Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com),
- *
- *  Derived from "arch/i386/kernel/process.c"
- *    Copyright (C) 1995, Linus Torvalds
- */
-
-/*
- * This file handles the architecture-dependent parts of process handling..
+ *    Copyright IBM Corp. 1999,2009
+ *    Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ *		 Hartmut Penner <hp@de.ibm.com>,
+ *		 Denis Joseph Barrow,
  */
 
 #include <linux/compiler.h>
@@ -168,34 +160,34 @@ void release_thread(struct task_struct *dead_task)
 }
 
 int copy_thread(int nr, unsigned long clone_flags, unsigned long new_stackp,
-	unsigned long unused,
-        struct task_struct * p, struct pt_regs * regs)
+		unsigned long unused,
+		struct task_struct *p, struct pt_regs *regs)
 {
-        struct fake_frame
-          {
-	    struct stack_frame sf;
-            struct pt_regs childregs;
-          } *frame;
-
-        frame = container_of(task_pt_regs(p), struct fake_frame, childregs);
-        p->thread.ksp = (unsigned long) frame;
+	struct fake_frame
+	{
+		struct stack_frame sf;
+		struct pt_regs childregs;
+	} *frame;
+
+	frame = container_of(task_pt_regs(p), struct fake_frame, childregs);
+	p->thread.ksp = (unsigned long) frame;
 	/* Store access registers to kernel stack of new process. */
-        frame->childregs = *regs;
+	frame->childregs = *regs;
 	frame->childregs.gprs[2] = 0;	/* child returns 0 on fork. */
-        frame->childregs.gprs[15] = new_stackp;
-        frame->sf.back_chain = 0;
+	frame->childregs.gprs[15] = new_stackp;
+	frame->sf.back_chain = 0;
 
-        /* new return point is ret_from_fork */
-        frame->sf.gprs[8] = (unsigned long) ret_from_fork;
+	/* new return point is ret_from_fork */
+	frame->sf.gprs[8] = (unsigned long) ret_from_fork;
 
-        /* fake return stack for resume(), don't go back to schedule */
-        frame->sf.gprs[9] = (unsigned long) frame;
+	/* fake return stack for resume(), don't go back to schedule */
+	frame->sf.gprs[9] = (unsigned long) frame;
 
 	/* Save access registers to new thread structure. */
 	save_access_regs(&p->thread.acrs[0]);
 
 #ifndef CONFIG_64BIT
-        /*
+	/*
 	 * save fprs to current->thread.fp_regs to merge them with
 	 * the emulated registers and then copy the result to the child.
 	 */
@@ -220,10 +212,9 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long new_stackp,
 #endif /* CONFIG_64BIT */
 	/* start new process with ar4 pointing to the correct address space */
 	p->thread.mm_segment = get_fs();
-        /* Don't copy debug registers */
-        memset(&p->thread.per_info,0,sizeof(p->thread.per_info));
-
-        return 0;
+	/* Don't copy debug registers */
+	memset(&p->thread.per_info, 0, sizeof(p->thread.per_info));
+	return 0;
 }
 
 SYSCALL_DEFINE0(fork)
@@ -311,7 +302,7 @@ out:
 int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs)
 {
 #ifndef CONFIG_64BIT
-        /*
+	/*
 	 * save fprs to current->thread.fp_regs to merge them with
 	 * the emulated registers and then copy the result to the dump.
 	 */
@@ -346,4 +337,3 @@ unsigned long get_wchan(struct task_struct *p)
 	}
 	return 0;
 }
-
-- 
cgit v1.2.3-55-g7522


From 5168ce2c647f02756803bef7b74906f485491a1c Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:23:53 +0100
Subject: [S390] cputime: initialize per thread timer values on fork

Initialize per thread timer values instead of just copying them from
the parent. That way it is easily possible to tell how much time a
thread spent in user/system context.
Doesn't fix a bug, this is just for debugging purposes.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/process.c | 5 +++++
 1 file changed, 5 insertions(+)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index e0563baa1cfe..e6b480625cb3 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -163,6 +163,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long new_stackp,
 		unsigned long unused,
 		struct task_struct *p, struct pt_regs *regs)
 {
+	struct thread_info *ti;
 	struct fake_frame
 	{
 		struct stack_frame sf;
@@ -214,6 +215,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long new_stackp,
 	p->thread.mm_segment = get_fs();
 	/* Don't copy debug registers */
 	memset(&p->thread.per_info, 0, sizeof(p->thread.per_info));
+	/* Initialize per thread user and system timer values */
+	ti = task_thread_info(p);
+	ti->user_timer = 0;
+	ti->system_timer = 0;
 	return 0;
 }
 
-- 
cgit v1.2.3-55-g7522


From 3324e60aafa9aeb1009878d45079b67367a5e2b6 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger
Date: Thu, 26 Mar 2009 15:23:56 +0100
Subject: [S390] lockdep: trace hardirq off in smp_send_stop

With lockdep we got the following trace after a panic:

Badness at /home/autobuild/BUILD/linux-2.6.28-20090204/kernel/lockdep.c:2878
[...]
Call Trace:
 [<0000000000176334>] lock_acquire+0x54/0xbc
 [<000000000050b4fe>] __atomic_notifier_call_chain+0x6e/0xdc
 [<000000000050b59c>] atomic_notifier_call_chain+0x30/0x44
 [<0000000000504274>] panic+0xd0/0x1e8
[...]
INFO: lockdep is turned off.
Last Breaking-Event-Address:
 [<0000000000170e62>] check_flags+0xae/0x15c
possible reason: unannotated irqs-off.

lockdep is right. We missed a trace_hardirq_off in our smp_send_stop
function and smp_send_stop is called before the panic call chain.

Reported-by: Mijo <Safradin mijo@linux.vnet.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/smp.c | 2 ++
 1 file changed, 2 insertions(+)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index e279d0fbbbe8..a8858634dd05 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -32,6 +32,7 @@
 #include <linux/delay.h>
 #include <linux/cache.h>
 #include <linux/interrupt.h>
+#include <linux/irqflags.h>
 #include <linux/cpu.h>
 #include <linux/timex.h>
 #include <linux/bootmem.h>
@@ -75,6 +76,7 @@ void smp_send_stop(void)
 
 	/* Disable all interrupts/machine checks */
 	__load_psw_mask(psw_kernel_bits & ~PSW_MASK_MCHECK);
+	trace_hardirqs_off();
 
 	/* stop all processors */
 	for_each_online_cpu(cpu) {
-- 
cgit v1.2.3-55-g7522


From 702d9e584feb028ed7e2a6d2b103b8ea57622ff2 Mon Sep 17 00:00:00 2001
From: Carsten Otte
Date: Thu, 26 Mar 2009 15:23:57 +0100
Subject: [S390] check addressing mode in s390_enable_sie

The sie instruction requires address spaces to be switched
to run proper. This patch verifies that this is the case
in s390_enable_sie, otherwise the kernel would crash badly
as soon as the process runs into sie.

Signed-off-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/mm/pgtable.c | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'arch/s390')

diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c
index 6b6ddc4ea02b..9bf86125f6f3 100644
--- a/arch/s390/mm/pgtable.c
+++ b/arch/s390/mm/pgtable.c
@@ -258,6 +258,10 @@ int s390_enable_sie(void)
 	struct task_struct *tsk = current;
 	struct mm_struct *mm, *old_mm;
 
+	/* Do we have switched amode? If no, we cannot do sie */
+	if (!switch_amode)
+		return -EINVAL;
+
 	/* Do we have pgstes? if yes, we are done */
 	if (tsk->mm->context.has_pgste)
 		return 0;
-- 
cgit v1.2.3-55-g7522


From 92e6ecf392fac3082653ac9d84b1bdf53d0ea160 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger
Date: Thu, 26 Mar 2009 15:23:58 +0100
Subject: [S390] Fix hypervisor detection for KVM

Currently we use the cpuid (via STIDP instruction) to recognize LPAR,
z/VM and KVM.
The architecture states, that bit 0-7 of STIDP returns all zero, and
if STIDP is executed in a virtual machine, the VM operating system
will replace bits 0-7 with FF.

KVM should not use FE to distinguish z/VM from KVM for interested
guests. The proper way to detect the hypervisor is the STSI (Store
System Information) instruction, which return information about the
hypervisors via function code 3, selector1=2, selector2=2.

This patch changes the detection routine of Linux to use STSI instead
of STIDP. This detection is earlier than bootmem, we have to use a
static buffer. Since STSI expects a 4kb block (4kb aligned) this
patch also changes the init.data alignment for s390. As this section
will be freed during boot, this should be no problem.

Patch is tested with LPAR, z/VM, KVM on LPAR, and KVM under z/VM.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/sysinfo.h |  1 +
 arch/s390/kernel/early.c        | 22 +++++++++++++---------
 arch/s390/kernel/vmlinux.lds.S  |  2 ++
 arch/s390/kvm/kvm-s390.c        |  2 +-
 4 files changed, 17 insertions(+), 10 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/sysinfo.h b/arch/s390/include/asm/sysinfo.h
index ad93212d9e16..9d70057d828c 100644
--- a/arch/s390/include/asm/sysinfo.h
+++ b/arch/s390/include/asm/sysinfo.h
@@ -100,6 +100,7 @@ struct sysinfo_3_2_2 {
 		char reserved_1[24];
 
 	} vm[8];
+	char reserved_544[3552];
 };
 
 static inline int stsi(void *sysinfo, int fc, int sel1, int sel2)
diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index 2a2ca268b1dd..d61eb3334edf 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -6,6 +6,7 @@
  *		 Heiko Carstens <heiko.carstens@de.ibm.com>
  */
 
+#include <linux/compiler.h>
 #include <linux/init.h>
 #include <linux/errno.h>
 #include <linux/string.h>
@@ -20,6 +21,7 @@
 #include <asm/processor.h>
 #include <asm/sections.h>
 #include <asm/setup.h>
+#include <asm/sysinfo.h>
 #include <asm/cpcmd.h>
 #include <asm/sclp.h>
 #include "entry.h"
@@ -173,19 +175,21 @@ static noinline __init void init_kernel_storage_key(void)
 		page_set_storage_key(init_pfn << PAGE_SHIFT, PAGE_DEFAULT_KEY);
 }
 
+static __initdata struct sysinfo_3_2_2 vmms __aligned(PAGE_SIZE);
+
 static noinline __init void detect_machine_type(void)
 {
-	struct cpuinfo_S390 *cpuinfo = &S390_lowcore.cpu_data;
-
-	get_cpu_id(&S390_lowcore.cpu_data.cpu_id);
-
-	/* Running under z/VM ? */
-	if (cpuinfo->cpu_id.version == 0xff)
-		machine_flags |= MACHINE_FLAG_VM;
+	/* No VM information? Looks like LPAR */
+	if (stsi(&vmms, 3, 2, 2) == -ENOSYS)
+		return;
+	if (!vmms.count)
+		return;
 
-	/* Running under KVM ? */
-	if (cpuinfo->cpu_id.version == 0xfe)
+	/* Running under KVM? If not we assume z/VM */
+	if (!memcmp(vmms.vm[0].cpi, "\xd2\xe5\xd4", 3))
 		machine_flags |= MACHINE_FLAG_KVM;
+	else
+		machine_flags |= MACHINE_FLAG_VM;
 }
 
 static __init void early_pgm_check_handler(void)
diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S
index d796d05c9c01..7a2063eb88f0 100644
--- a/arch/s390/kernel/vmlinux.lds.S
+++ b/arch/s390/kernel/vmlinux.lds.S
@@ -108,6 +108,8 @@ SECTIONS
 		EXIT_TEXT
 	}
 
+	/* early.c uses stsi, which requires page aligned data. */
+	. = ALIGN(PAGE_SIZE);
 	.init.data : {
 		INIT_DATA
 	}
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 0d33893e1e89..c55a4b9ffd88 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -286,7 +286,7 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
 	setup_timer(&vcpu->arch.ckc_timer, kvm_s390_idle_wakeup,
 		 (unsigned long) vcpu);
 	get_cpu_id(&vcpu->arch.cpu_id);
-	vcpu->arch.cpu_id.version = 0xfe;
+	vcpu->arch.cpu_id.version = 0xff;
 	return 0;
 }
 
-- 
cgit v1.2.3-55-g7522


From 70193af9188113c9b4ff3dde1aed9f9c8f7c4f93 Mon Sep 17 00:00:00 2001
From: Sachin Sant
Date: Thu, 26 Mar 2009 15:24:00 +0100
Subject: [S390] Fix appldata build break with !NET

With CONFIG_NET not set appldata build breaks on s390.

arch/s390/appldata/built-in.o: In function appldata_get_net_sum_data:
appldata_net_sum.c:(.text+0x2684): undefined reference to dev_get_stats
appldata_net_sum.c:(.text+0x2688): undefined reference to init_net
appldata_net_sum.c:(.text+0x268c): undefined reference to init_net
appldata_net_sum.c:(.text+0x2694): undefined reference to dev_base_lock

The following patch fixes the issue for me.

Signed-off-by: Sachin Sant <sachinp@in.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/s390')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 6b0a3538dc63..3f470604a89a 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -521,7 +521,7 @@ config APPLDATA_OS
 
 config APPLDATA_NET_SUM
 	tristate "Monitor overall network statistics"
-	depends on APPLDATA_BASE
+	depends on APPLDATA_BASE && NET
 	help
 	  This provides network related data to the Linux - VM Monitor Stream,
 	  currently there is only a total sum of network I/O statistics, no
-- 
cgit v1.2.3-55-g7522


From f5daba1d4116d964435ddd99f32b6c80448a496b Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:01 +0100
Subject: [S390] split/move machine check handler code

Split machine check handler code and move it to cio and kernel code
where it belongs to. No functional change.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/crw.h |  68 ++++++
 arch/s390/include/asm/nmi.h |  66 ++++++
 arch/s390/kernel/Makefile   |   2 +-
 arch/s390/kernel/nmi.c      | 383 +++++++++++++++++++++++++++++++
 arch/s390/kernel/process.c  |   2 +-
 arch/s390/kvm/kvm-s390.c    |   4 +-
 drivers/s390/Makefile       |   1 -
 drivers/s390/cio/Makefile   |   2 +-
 drivers/s390/cio/chp.c      |   6 +-
 drivers/s390/cio/chsc.c     |   6 +-
 drivers/s390/cio/cio.c      |   3 +-
 drivers/s390/cio/crw.c      | 145 ++++++++++++
 drivers/s390/cio/css.c      |   6 +-
 drivers/s390/s390mach.c     | 537 --------------------------------------------
 drivers/s390/s390mach.h     | 122 ----------
 15 files changed, 677 insertions(+), 676 deletions(-)
 create mode 100644 arch/s390/include/asm/crw.h
 create mode 100644 arch/s390/include/asm/nmi.h
 create mode 100644 arch/s390/kernel/nmi.c
 create mode 100644 drivers/s390/cio/crw.c
 delete mode 100644 drivers/s390/s390mach.c
 delete mode 100644 drivers/s390/s390mach.h

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/crw.h b/arch/s390/include/asm/crw.h
new file mode 100644
index 000000000000..2185a6d619d3
--- /dev/null
+++ b/arch/s390/include/asm/crw.h
@@ -0,0 +1,68 @@
+/*
+ *   Data definitions for channel report processing
+ *    Copyright IBM Corp. 2000,2009
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>,
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>,
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>,
+ */
+
+#ifndef _ASM_S390_CRW_H
+#define _ASM_S390_CRW_H
+
+#include <linux/types.h>
+
+/*
+ * Channel Report Word
+ */
+struct crw {
+	__u32 res1 :  1;   /* reserved zero */
+	__u32 slct :  1;   /* solicited */
+	__u32 oflw :  1;   /* overflow */
+	__u32 chn  :  1;   /* chained */
+	__u32 rsc  :  4;   /* reporting source code */
+	__u32 anc  :  1;   /* ancillary report */
+	__u32 res2 :  1;   /* reserved zero */
+	__u32 erc  :  6;   /* error-recovery code */
+	__u32 rsid : 16;   /* reporting-source ID */
+} __attribute__ ((packed));
+
+typedef void (*crw_handler_t)(struct crw *, struct crw *, int);
+
+extern int crw_register_handler(int rsc, crw_handler_t handler);
+extern void crw_unregister_handler(int rsc);
+extern void crw_handle_channel_report(void);
+
+#define NR_RSCS 16
+
+#define CRW_RSC_MONITOR  0x2  /* monitoring facility */
+#define CRW_RSC_SCH	 0x3  /* subchannel */
+#define CRW_RSC_CPATH	 0x4  /* channel path */
+#define CRW_RSC_CONFIG	 0x9  /* configuration-alert facility */
+#define CRW_RSC_CSS	 0xB  /* channel subsystem */
+
+#define CRW_ERC_EVENT	 0x00 /* event information pending */
+#define CRW_ERC_AVAIL	 0x01 /* available */
+#define CRW_ERC_INIT	 0x02 /* initialized */
+#define CRW_ERC_TERROR	 0x03 /* temporary error */
+#define CRW_ERC_IPARM	 0x04 /* installed parm initialized */
+#define CRW_ERC_TERM	 0x05 /* terminal */
+#define CRW_ERC_PERRN	 0x06 /* perm. error, fac. not init */
+#define CRW_ERC_PERRI	 0x07 /* perm. error, facility init */
+#define CRW_ERC_PMOD	 0x08 /* installed parameters modified */
+
+static inline int stcrw(struct crw *pcrw)
+{
+	int ccode;
+
+	asm volatile(
+		"	stcrw	0(%2)\n"
+		"	ipm	%0\n"
+		"	srl	%0,28\n"
+		: "=d" (ccode), "=m" (*pcrw)
+		: "a" (pcrw)
+		: "cc" );
+	return ccode;
+}
+
+#endif /* _ASM_S390_CRW_H */
diff --git a/arch/s390/include/asm/nmi.h b/arch/s390/include/asm/nmi.h
new file mode 100644
index 000000000000..f4b60441adca
--- /dev/null
+++ b/arch/s390/include/asm/nmi.h
@@ -0,0 +1,66 @@
+/*
+ *   Machine check handler definitions
+ *
+ *    Copyright IBM Corp. 2000,2009
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>,
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>,
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>,
+ */
+
+#ifndef _ASM_S390_NMI_H
+#define _ASM_S390_NMI_H
+
+#include <linux/types.h>
+
+struct mci {
+	__u32 sd :  1; /* 00 system damage */
+	__u32 pd :  1; /* 01 instruction-processing damage */
+	__u32 sr :  1; /* 02 system recovery */
+	__u32	 :  1; /* 03 */
+	__u32 cd :  1; /* 04 timing-facility damage */
+	__u32 ed :  1; /* 05 external damage */
+	__u32	 :  1; /* 06 */
+	__u32 dg :  1; /* 07 degradation */
+	__u32 w  :  1; /* 08 warning pending */
+	__u32 cp :  1; /* 09 channel-report pending */
+	__u32 sp :  1; /* 10 service-processor damage */
+	__u32 ck :  1; /* 11 channel-subsystem damage */
+	__u32	 :  2; /* 12-13 */
+	__u32 b  :  1; /* 14 backed up */
+	__u32	 :  1; /* 15 */
+	__u32 se :  1; /* 16 storage error uncorrected */
+	__u32 sc :  1; /* 17 storage error corrected */
+	__u32 ke :  1; /* 18 storage-key error uncorrected */
+	__u32 ds :  1; /* 19 storage degradation */
+	__u32 wp :  1; /* 20 psw mwp validity */
+	__u32 ms :  1; /* 21 psw mask and key validity */
+	__u32 pm :  1; /* 22 psw program mask and cc validity */
+	__u32 ia :  1; /* 23 psw instruction address validity */
+	__u32 fa :  1; /* 24 failing storage address validity */
+	__u32	 :  1; /* 25 */
+	__u32 ec :  1; /* 26 external damage code validity */
+	__u32 fp :  1; /* 27 floating point register validity */
+	__u32 gr :  1; /* 28 general register validity */
+	__u32 cr :  1; /* 29 control register validity */
+	__u32	 :  1; /* 30 */
+	__u32 st :  1; /* 31 storage logical validity */
+	__u32 ie :  1; /* 32 indirect storage error */
+	__u32 ar :  1; /* 33 access register validity */
+	__u32 da :  1; /* 34 delayed access exception */
+	__u32	 :  7; /* 35-41 */
+	__u32 pr :  1; /* 42 tod programmable register validity */
+	__u32 fc :  1; /* 43 fp control register validity */
+	__u32 ap :  1; /* 44 ancillary report */
+	__u32	 :  1; /* 45 */
+	__u32 ct :  1; /* 46 cpu timer validity */
+	__u32 cc :  1; /* 47 clock comparator validity */
+	__u32	 : 16; /* 47-63 */
+};
+
+struct pt_regs;
+
+extern void s390_handle_mcck(void);
+extern void s390_do_machine_check(struct pt_regs *regs);
+
+#endif /* _ASM_S390_NMI_H */
diff --git a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
index 33e7aee70513..228e3105ded7 100644
--- a/arch/s390/kernel/Makefile
+++ b/arch/s390/kernel/Makefile
@@ -22,7 +22,7 @@ CFLAGS_sysinfo.o += -Iinclude/math-emu -Iarch/s390/math-emu -w
 obj-y	:=  bitmap.o traps.o time.o process.o base.o early.o setup.o \
 	    processor.o sys_s390.o ptrace.o signal.o cpcmd.o ebcdic.o \
 	    s390_ext.o debug.o irq.o ipl.o dis.o diag.o mem_detect.o \
-	    vdso.o vtime.o sysinfo.o
+	    vdso.o vtime.o sysinfo.o nmi.o
 
 obj-y	+= $(if $(CONFIG_64BIT),entry64.o,entry.o)
 obj-y	+= $(if $(CONFIG_64BIT),reipl64.o,reipl.o)
diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c
new file mode 100644
index 000000000000..8d50eaf76a60
--- /dev/null
+++ b/arch/s390/kernel/nmi.c
@@ -0,0 +1,383 @@
+/*
+ *   Machine check handler
+ *
+ *    Copyright IBM Corp. 2000,2009
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>,
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>,
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>,
+ */
+
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/time.h>
+#include <linux/module.h>
+#include <asm/lowcore.h>
+#include <asm/smp.h>
+#include <asm/etr.h>
+#include <asm/cpu.h>
+#include <asm/nmi.h>
+#include <asm/crw.h>
+
+struct mcck_struct {
+	int kill_task;
+	int channel_report;
+	int warning;
+	unsigned long long mcck_code;
+};
+
+static DEFINE_PER_CPU(struct mcck_struct, cpu_mcck);
+
+static NORET_TYPE void s390_handle_damage(char *msg)
+{
+	smp_send_stop();
+	disabled_wait((unsigned long) __builtin_return_address(0));
+	while (1);
+}
+
+/*
+ * Main machine check handler function. Will be called with interrupts enabled
+ * or disabled and machine checks enabled or disabled.
+ */
+void s390_handle_mcck(void)
+{
+	unsigned long flags;
+	struct mcck_struct mcck;
+
+	/*
+	 * Disable machine checks and get the current state of accumulated
+	 * machine checks. Afterwards delete the old state and enable machine
+	 * checks again.
+	 */
+	local_irq_save(flags);
+	local_mcck_disable();
+	mcck = __get_cpu_var(cpu_mcck);
+	memset(&__get_cpu_var(cpu_mcck), 0, sizeof(struct mcck_struct));
+	clear_thread_flag(TIF_MCCK_PENDING);
+	local_mcck_enable();
+	local_irq_restore(flags);
+
+	if (mcck.channel_report)
+		crw_handle_channel_report();
+
+#ifdef CONFIG_MACHCHK_WARNING
+/*
+ * The warning may remain for a prolonged period on the bare iron.
+ * (actually till the machine is powered off, or until the problem is gone)
+ * So we just stop listening for the WARNING MCH and prevent continuously
+ * being interrupted.  One caveat is however, that we must do this per
+ * processor and cannot use the smp version of ctl_clear_bit().
+ * On VM we only get one interrupt per virtally presented machinecheck.
+ * Though one suffices, we may get one interrupt per (virtual) processor.
+ */
+	if (mcck.warning) {	/* WARNING pending ? */
+		static int mchchk_wng_posted = 0;
+		/*
+		 * Use single machine clear, as we cannot handle smp right now
+		 */
+		__ctl_clear_bit(14, 24);	/* Disable WARNING MCH */
+		if (xchg(&mchchk_wng_posted, 1) == 0)
+			kill_cad_pid(SIGPWR, 1);
+	}
+#endif
+
+	if (mcck.kill_task) {
+		local_irq_enable();
+		printk(KERN_EMERG "mcck: Terminating task because of machine "
+		       "malfunction (code 0x%016llx).\n", mcck.mcck_code);
+		printk(KERN_EMERG "mcck: task: %s, pid: %d.\n",
+		       current->comm, current->pid);
+		do_exit(SIGSEGV);
+	}
+}
+EXPORT_SYMBOL_GPL(s390_handle_mcck);
+
+/*
+ * returns 0 if all registers could be validated
+ * returns 1 otherwise
+ */
+static int notrace s390_revalidate_registers(struct mci *mci)
+{
+	int kill_task;
+	u64 tmpclock;
+	u64 zero;
+	void *fpt_save_area, *fpt_creg_save_area;
+
+	kill_task = 0;
+	zero = 0;
+
+	if (!mci->gr) {
+		/*
+		 * General purpose registers couldn't be restored and have
+		 * unknown contents. Process needs to be terminated.
+		 */
+		kill_task = 1;
+	}
+	if (!mci->fp) {
+		/*
+		 * Floating point registers can't be restored and
+		 * therefore the process needs to be terminated.
+		 */
+		kill_task = 1;
+	}
+#ifndef CONFIG_64BIT
+	asm volatile(
+		"	ld	0,0(%0)\n"
+		"	ld	2,8(%0)\n"
+		"	ld	4,16(%0)\n"
+		"	ld	6,24(%0)"
+		: : "a" (&S390_lowcore.floating_pt_save_area));
+#endif
+
+	if (MACHINE_HAS_IEEE) {
+#ifdef CONFIG_64BIT
+		fpt_save_area = &S390_lowcore.floating_pt_save_area;
+		fpt_creg_save_area = &S390_lowcore.fpt_creg_save_area;
+#else
+		fpt_save_area = (void *) S390_lowcore.extended_save_area_addr;
+		fpt_creg_save_area = fpt_save_area + 128;
+#endif
+		if (!mci->fc) {
+			/*
+			 * Floating point control register can't be restored.
+			 * Task will be terminated.
+			 */
+			asm volatile("lfpc 0(%0)" : : "a" (&zero), "m" (zero));
+			kill_task = 1;
+
+		} else
+			asm volatile("lfpc 0(%0)" : : "a" (fpt_creg_save_area));
+
+		asm volatile(
+			"	ld	0,0(%0)\n"
+			"	ld	1,8(%0)\n"
+			"	ld	2,16(%0)\n"
+			"	ld	3,24(%0)\n"
+			"	ld	4,32(%0)\n"
+			"	ld	5,40(%0)\n"
+			"	ld	6,48(%0)\n"
+			"	ld	7,56(%0)\n"
+			"	ld	8,64(%0)\n"
+			"	ld	9,72(%0)\n"
+			"	ld	10,80(%0)\n"
+			"	ld	11,88(%0)\n"
+			"	ld	12,96(%0)\n"
+			"	ld	13,104(%0)\n"
+			"	ld	14,112(%0)\n"
+			"	ld	15,120(%0)\n"
+			: : "a" (fpt_save_area));
+	}
+	/* Revalidate access registers */
+	asm volatile(
+		"	lam	0,15,0(%0)"
+		: : "a" (&S390_lowcore.access_regs_save_area));
+	if (!mci->ar) {
+		/*
+		 * Access registers have unknown contents.
+		 * Terminating task.
+		 */
+		kill_task = 1;
+	}
+	/* Revalidate control registers */
+	if (!mci->cr) {
+		/*
+		 * Control registers have unknown contents.
+		 * Can't recover and therefore stopping machine.
+		 */
+		s390_handle_damage("invalid control registers.");
+	} else {
+#ifdef CONFIG_64BIT
+		asm volatile(
+			"	lctlg	0,15,0(%0)"
+			: : "a" (&S390_lowcore.cregs_save_area));
+#else
+		asm volatile(
+			"	lctl	0,15,0(%0)"
+			: : "a" (&S390_lowcore.cregs_save_area));
+#endif
+	}
+	/*
+	 * We don't even try to revalidate the TOD register, since we simply
+	 * can't write something sensible into that register.
+	 */
+#ifdef CONFIG_64BIT
+	/*
+	 * See if we can revalidate the TOD programmable register with its
+	 * old contents (should be zero) otherwise set it to zero.
+	 */
+	if (!mci->pr)
+		asm volatile(
+			"	sr	0,0\n"
+			"	sckpf"
+			: : : "0", "cc");
+	else
+		asm volatile(
+			"	l	0,0(%0)\n"
+			"	sckpf"
+			: : "a" (&S390_lowcore.tod_progreg_save_area)
+			: "0", "cc");
+#endif
+	/* Revalidate clock comparator register */
+	asm volatile(
+		"	stck	0(%1)\n"
+		"	sckc	0(%1)"
+		: "=m" (tmpclock) : "a" (&(tmpclock)) : "cc", "memory");
+
+	/* Check if old PSW is valid */
+	if (!mci->wp)
+		/*
+		 * Can't tell if we come from user or kernel mode
+		 * -> stopping machine.
+		 */
+		s390_handle_damage("old psw invalid.");
+
+	if (!mci->ms || !mci->pm || !mci->ia)
+		kill_task = 1;
+
+	return kill_task;
+}
+
+#define MAX_IPD_COUNT	29
+#define MAX_IPD_TIME	(5 * 60 * USEC_PER_SEC) /* 5 minutes */
+
+#define ED_STP_ISLAND	6	/* External damage STP island check */
+#define ED_STP_SYNC	7	/* External damage STP sync check */
+#define ED_ETR_SYNC	12	/* External damage ETR sync check */
+#define ED_ETR_SWITCH	13	/* External damage ETR switch to local */
+
+/*
+ * machine check handler.
+ */
+void notrace s390_do_machine_check(struct pt_regs *regs)
+{
+	static int ipd_count;
+	static DEFINE_SPINLOCK(ipd_lock);
+	static unsigned long long last_ipd;
+	struct mcck_struct *mcck;
+	unsigned long long tmp;
+	struct mci *mci;
+	int umode;
+
+	lockdep_off();
+	s390_idle_check();
+
+	mci = (struct mci *) &S390_lowcore.mcck_interruption_code;
+	mcck = &__get_cpu_var(cpu_mcck);
+	umode = user_mode(regs);
+
+	if (mci->sd) {
+		/* System damage -> stopping machine */
+		s390_handle_damage("received system damage machine check.");
+	}
+	if (mci->pd) {
+		if (mci->b) {
+			/* Processing backup -> verify if we can survive this */
+			u64 z_mcic, o_mcic, t_mcic;
+#ifdef CONFIG_64BIT
+			z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<29);
+			o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 |
+				  1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 |
+				  1ULL<<30 | 1ULL<<21 | 1ULL<<20 | 1ULL<<17 |
+				  1ULL<<16);
+#else
+			z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<57 | 1ULL<<50 |
+				  1ULL<<29);
+			o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 |
+				  1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 |
+				  1ULL<<30 | 1ULL<<20 | 1ULL<<17 | 1ULL<<16);
+#endif
+			t_mcic = *(u64 *)mci;
+
+			if (((t_mcic & z_mcic) != 0) ||
+			    ((t_mcic & o_mcic) != o_mcic)) {
+				s390_handle_damage("processing backup machine "
+						   "check with damage.");
+			}
+
+			/*
+			 * Nullifying exigent condition, therefore we might
+			 * retry this instruction.
+			 */
+			spin_lock(&ipd_lock);
+			tmp = get_clock();
+			if (((tmp - last_ipd) >> 12) < MAX_IPD_TIME)
+				ipd_count++;
+			else
+				ipd_count = 1;
+			last_ipd = tmp;
+			if (ipd_count == MAX_IPD_COUNT)
+				s390_handle_damage("too many ipd retries.");
+			spin_unlock(&ipd_lock);
+		} else {
+			/* Processing damage -> stopping machine */
+			s390_handle_damage("received instruction processing "
+					   "damage machine check.");
+		}
+	}
+	if (s390_revalidate_registers(mci)) {
+		if (umode) {
+			/*
+			 * Couldn't restore all register contents while in
+			 * user mode -> mark task for termination.
+			 */
+			mcck->kill_task = 1;
+			mcck->mcck_code = *(unsigned long long *) mci;
+			set_thread_flag(TIF_MCCK_PENDING);
+		} else {
+			/*
+			 * Couldn't restore all register contents while in
+			 * kernel mode -> stopping machine.
+			 */
+			s390_handle_damage("unable to revalidate registers.");
+		}
+	}
+	if (mci->cd) {
+		/* Timing facility damage */
+		s390_handle_damage("TOD clock damaged");
+	}
+	if (mci->ed && mci->ec) {
+		/* External damage */
+		if (S390_lowcore.external_damage_code & (1U << ED_ETR_SYNC))
+			etr_sync_check();
+		if (S390_lowcore.external_damage_code & (1U << ED_ETR_SWITCH))
+			etr_switch_to_local();
+		if (S390_lowcore.external_damage_code & (1U << ED_STP_SYNC))
+			stp_sync_check();
+		if (S390_lowcore.external_damage_code & (1U << ED_STP_ISLAND))
+			stp_island_check();
+	}
+	if (mci->se)
+		/* Storage error uncorrected */
+		s390_handle_damage("received storage error uncorrected "
+				   "machine check.");
+	if (mci->ke)
+		/* Storage key-error uncorrected */
+		s390_handle_damage("received storage key-error uncorrected "
+				   "machine check.");
+	if (mci->ds && mci->fa)
+		/* Storage degradation */
+		s390_handle_damage("received storage degradation machine "
+				   "check.");
+	if (mci->cp) {
+		/* Channel report word pending */
+		mcck->channel_report = 1;
+		set_thread_flag(TIF_MCCK_PENDING);
+	}
+	if (mci->w) {
+		/* Warning pending */
+		mcck->warning = 1;
+		set_thread_flag(TIF_MCCK_PENDING);
+	}
+	lockdep_on();
+}
+
+static int __init machine_check_init(void)
+{
+	ctl_set_bit(14, 25);	/* enable external damage MCH */
+	ctl_set_bit(14, 27);	/* enable system recovery MCH */
+#ifdef CONFIG_MACHCHK_WARNING
+	ctl_set_bit(14, 24);	/* enable warning MCH */
+#endif
+	return 0;
+}
+arch_initcall(machine_check_init);
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index e6b480625cb3..bd616fb31e75 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -39,6 +39,7 @@
 #include <asm/processor.h>
 #include <asm/irq.h>
 #include <asm/timer.h>
+#include <asm/nmi.h>
 #include "entry.h"
 
 asmlinkage void ret_from_fork(void) asm ("ret_from_fork");
@@ -68,7 +69,6 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
 	return sf->gprs[8];
 }
 
-extern void s390_handle_mcck(void);
 /*
  * The idle loop on a S390...
  */
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index c55a4b9ffd88..caa4d2877016 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -23,7 +23,7 @@
 #include <linux/timer.h>
 #include <asm/lowcore.h>
 #include <asm/pgtable.h>
-
+#include <asm/nmi.h>
 #include "kvm-s390.h"
 #include "gaccess.h"
 
@@ -440,8 +440,6 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
 	return -EINVAL; /* not implemented yet */
 }
 
-extern void s390_handle_mcck(void);
-
 static void __vcpu_run(struct kvm_vcpu *vcpu)
 {
 	memcpy(&vcpu->arch.sie_block->gg14, &vcpu->arch.guest_gprs[14], 16);
diff --git a/drivers/s390/Makefile b/drivers/s390/Makefile
index 0828dc839355..95bccfd3f169 100644
--- a/drivers/s390/Makefile
+++ b/drivers/s390/Makefile
@@ -2,7 +2,6 @@
 # Makefile for the S/390 specific device drivers
 #
 
-obj-y += s390mach.o
 obj-y += cio/ block/ char/ crypto/ net/ scsi/ kvm/
 
 drivers-y += drivers/s390/built-in.o
diff --git a/drivers/s390/cio/Makefile b/drivers/s390/cio/Makefile
index bd79bd165396..adb3dd301528 100644
--- a/drivers/s390/cio/Makefile
+++ b/drivers/s390/cio/Makefile
@@ -3,7 +3,7 @@
 #
 
 obj-y += airq.o blacklist.o chsc.o cio.o css.o chp.o idset.o isc.o scsw.o \
-	fcx.o itcw.o
+	fcx.o itcw.o crw.o
 ccw_device-objs += device.o device_fsm.o device_ops.o
 ccw_device-objs += device_id.o device_pgid.o device_status.o
 obj-y += ccw_device.o cmf.o
diff --git a/drivers/s390/cio/chp.c b/drivers/s390/cio/chp.c
index 1246f61a5338..3e5f304ad88f 100644
--- a/drivers/s390/cio/chp.c
+++ b/drivers/s390/cio/chp.c
@@ -17,8 +17,8 @@
 #include <linux/errno.h>
 #include <asm/chpid.h>
 #include <asm/sclp.h>
+#include <asm/crw.h>
 
-#include "../s390mach.h"
 #include "cio.h"
 #include "css.h"
 #include "ioasm.h"
@@ -706,12 +706,12 @@ static int __init chp_init(void)
 	struct chp_id chpid;
 	int ret;
 
-	ret = s390_register_crw_handler(CRW_RSC_CPATH, chp_process_crw);
+	ret = crw_register_handler(CRW_RSC_CPATH, chp_process_crw);
 	if (ret)
 		return ret;
 	chp_wq = create_singlethread_workqueue("cio_chp");
 	if (!chp_wq) {
-		s390_unregister_crw_handler(CRW_RSC_CPATH);
+		crw_unregister_handler(CRW_RSC_CPATH);
 		return -ENOMEM;
 	}
 	INIT_WORK(&cfg_work, cfg_func);
diff --git a/drivers/s390/cio/chsc.c b/drivers/s390/cio/chsc.c
index ebab6ea4659b..7399b07a1aeb 100644
--- a/drivers/s390/cio/chsc.c
+++ b/drivers/s390/cio/chsc.c
@@ -19,8 +19,8 @@
 #include <asm/cio.h>
 #include <asm/chpid.h>
 #include <asm/chsc.h>
+#include <asm/crw.h>
 
-#include "../s390mach.h"
 #include "css.h"
 #include "cio.h"
 #include "cio_debug.h"
@@ -820,7 +820,7 @@ int __init chsc_alloc_sei_area(void)
 			      "chsc machine checks!\n");
 		return -ENOMEM;
 	}
-	ret = s390_register_crw_handler(CRW_RSC_CSS, chsc_process_crw);
+	ret = crw_register_handler(CRW_RSC_CSS, chsc_process_crw);
 	if (ret)
 		kfree(sei_page);
 	return ret;
@@ -828,7 +828,7 @@ int __init chsc_alloc_sei_area(void)
 
 void __init chsc_free_sei_area(void)
 {
-	s390_unregister_crw_handler(CRW_RSC_CSS);
+	crw_unregister_handler(CRW_RSC_CSS);
 	kfree(sei_page);
 }
 
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c
index 659f8a791656..73135c5e9dfb 100644
--- a/drivers/s390/cio/cio.c
+++ b/drivers/s390/cio/cio.c
@@ -30,6 +30,8 @@
 #include <asm/isc.h>
 #include <asm/cpu.h>
 #include <asm/fcx.h>
+#include <asm/nmi.h>
+#include <asm/crw.h>
 #include "cio.h"
 #include "css.h"
 #include "chsc.h"
@@ -38,7 +40,6 @@
 #include "blacklist.h"
 #include "cio_debug.h"
 #include "chp.h"
-#include "../s390mach.h"
 
 debug_info_t *cio_debug_msg_id;
 debug_info_t *cio_debug_trace_id;
diff --git a/drivers/s390/cio/crw.c b/drivers/s390/cio/crw.c
new file mode 100644
index 000000000000..508f88f6420c
--- /dev/null
+++ b/drivers/s390/cio/crw.c
@@ -0,0 +1,145 @@
+/*
+ *   Channel report handling code
+ *
+ *    Copyright IBM Corp. 2000,2009
+ *    Author(s): Ingo Adlung <adlung@de.ibm.com>,
+ *		 Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ *		 Cornelia Huck <cornelia.huck@de.ibm.com>,
+ *		 Heiko Carstens <heiko.carstens@de.ibm.com>,
+ */
+
+#include <linux/semaphore.h>
+#include <linux/kthread.h>
+#include <linux/init.h>
+#include <asm/crw.h>
+
+static struct semaphore crw_semaphore;
+static crw_handler_t crw_handlers[NR_RSCS];
+
+/**
+ * crw_register_handler() - register a channel report word handler
+ * @rsc: reporting source code to handle
+ * @handler: handler to be registered
+ *
+ * Returns %0 on success and a negative error value otherwise.
+ */
+int crw_register_handler(int rsc, crw_handler_t handler)
+{
+	if ((rsc < 0) || (rsc >= NR_RSCS))
+		return -EINVAL;
+	if (!cmpxchg(&crw_handlers[rsc], NULL, handler))
+		return 0;
+	return -EBUSY;
+}
+
+/**
+ * crw_unregister_handler() - unregister a channel report word handler
+ * @rsc: reporting source code to handle
+ */
+void crw_unregister_handler(int rsc)
+{
+	if ((rsc < 0) || (rsc >= NR_RSCS))
+		return;
+	xchg(&crw_handlers[rsc], NULL);
+	synchronize_sched();
+}
+
+/*
+ * Retrieve CRWs and call function to handle event.
+ */
+static int crw_collect_info(void *unused)
+{
+	struct crw crw[2];
+	int ccode;
+	unsigned int chain;
+	int ignore;
+
+repeat:
+	ignore = down_interruptible(&crw_semaphore);
+	chain = 0;
+	while (1) {
+		if (unlikely(chain > 1)) {
+			struct crw tmp_crw;
+
+			printk(KERN_WARNING"%s: Code does not support more "
+			       "than two chained crws; please report to "
+			       "linux390@de.ibm.com!\n", __func__);
+			ccode = stcrw(&tmp_crw);
+			printk(KERN_WARNING"%s: crw reports slct=%d, oflw=%d, "
+			       "chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
+			       __func__, tmp_crw.slct, tmp_crw.oflw,
+			       tmp_crw.chn, tmp_crw.rsc, tmp_crw.anc,
+			       tmp_crw.erc, tmp_crw.rsid);
+			printk(KERN_WARNING"%s: This was crw number %x in the "
+			       "chain\n", __func__, chain);
+			if (ccode != 0)
+				break;
+			chain = tmp_crw.chn ? chain + 1 : 0;
+			continue;
+		}
+		ccode = stcrw(&crw[chain]);
+		if (ccode != 0)
+			break;
+		printk(KERN_DEBUG "crw_info : CRW reports slct=%d, oflw=%d, "
+		       "chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
+		       crw[chain].slct, crw[chain].oflw, crw[chain].chn,
+		       crw[chain].rsc, crw[chain].anc, crw[chain].erc,
+		       crw[chain].rsid);
+		/* Check for overflows. */
+		if (crw[chain].oflw) {
+			int i;
+
+			pr_debug("%s: crw overflow detected!\n", __func__);
+			for (i = 0; i < NR_RSCS; i++) {
+				if (crw_handlers[i])
+					crw_handlers[i](NULL, NULL, 1);
+			}
+			chain = 0;
+			continue;
+		}
+		if (crw[0].chn && !chain) {
+			chain++;
+			continue;
+		}
+		if (crw_handlers[crw[chain].rsc])
+			crw_handlers[crw[chain].rsc](&crw[0],
+						     chain ? &crw[1] : NULL,
+						     0);
+		/* chain is always 0 or 1 here. */
+		chain = crw[chain].chn ? chain + 1 : 0;
+	}
+	goto repeat;
+	return 0;
+}
+
+void crw_handle_channel_report(void)
+{
+	up(&crw_semaphore);
+}
+
+/*
+ * Separate initcall needed for semaphore initialization since
+ * crw_handle_channel_report might be called before crw_machine_check_init.
+ */
+static int __init crw_init_semaphore(void)
+{
+	init_MUTEX_LOCKED(&crw_semaphore);
+	return 0;
+}
+pure_initcall(crw_init_semaphore);
+
+/*
+ * Machine checks for the channel subsystem must be enabled
+ * after the channel subsystem is initialized
+ */
+static int __init crw_machine_check_init(void)
+{
+	struct task_struct *task;
+
+	task = kthread_run(crw_collect_info, NULL, "kmcheck");
+	if (IS_ERR(task))
+		return PTR_ERR(task);
+	ctl_set_bit(14, 28);	/* enable channel report MCH */
+	return 0;
+}
+device_initcall(crw_machine_check_init);
diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c
index 8019288bc6de..a5fc56371ba8 100644
--- a/drivers/s390/cio/css.c
+++ b/drivers/s390/cio/css.c
@@ -18,8 +18,8 @@
 #include <linux/list.h>
 #include <linux/reboot.h>
 #include <asm/isc.h>
+#include <asm/crw.h>
 
-#include "../s390mach.h"
 #include "css.h"
 #include "cio.h"
 #include "cio_debug.h"
@@ -765,7 +765,7 @@ init_channel_subsystem (void)
 	if (ret)
 		goto out;
 
-	ret = s390_register_crw_handler(CRW_RSC_SCH, css_process_crw);
+	ret = crw_register_handler(CRW_RSC_SCH, css_process_crw);
 	if (ret)
 		goto out;
 
@@ -845,7 +845,7 @@ out_unregister:
 out_bus:
 	bus_unregister(&css_bus_type);
 out:
-	s390_unregister_crw_handler(CRW_RSC_CSS);
+	crw_unregister_handler(CRW_RSC_CSS);
 	chsc_free_sei_area();
 	kfree(slow_subchannel_set);
 	pr_alert("The CSS device driver initialization failed with "
diff --git a/drivers/s390/s390mach.c b/drivers/s390/s390mach.c
deleted file mode 100644
index 8e437052e076..000000000000
--- a/drivers/s390/s390mach.c
+++ /dev/null
@@ -1,537 +0,0 @@
-/*
- *  drivers/s390/s390mach.c
- *   S/390 machine check handler
- *
- *    Copyright IBM Corp. 2000,2008
- *    Author(s): Ingo Adlung (adlung@de.ibm.com)
- *		 Martin Schwidefsky (schwidefsky@de.ibm.com)
- *		 Cornelia Huck <cornelia.huck@de.ibm.com>
- */
-
-#include <linux/init.h>
-#include <linux/sched.h>
-#include <linux/errno.h>
-#include <linux/workqueue.h>
-#include <linux/time.h>
-#include <linux/device.h>
-#include <linux/kthread.h>
-#include <asm/etr.h>
-#include <asm/lowcore.h>
-#include <asm/cio.h>
-#include <asm/cpu.h>
-#include "s390mach.h"
-
-static struct semaphore m_sem;
-
-static NORET_TYPE void
-s390_handle_damage(char *msg)
-{
-#ifdef CONFIG_SMP
-	smp_send_stop();
-#endif
-	disabled_wait((unsigned long) __builtin_return_address(0));
-	for(;;);
-}
-
-static crw_handler_t crw_handlers[NR_RSCS];
-
-/**
- * s390_register_crw_handler() - register a channel report word handler
- * @rsc: reporting source code to handle
- * @handler: handler to be registered
- *
- * Returns %0 on success and a negative error value otherwise.
- */
-int s390_register_crw_handler(int rsc, crw_handler_t handler)
-{
-	if ((rsc < 0) || (rsc >= NR_RSCS))
-		return -EINVAL;
-	if (!cmpxchg(&crw_handlers[rsc], NULL, handler))
-		return 0;
-	return -EBUSY;
-}
-
-/**
- * s390_unregister_crw_handler() - unregister a channel report word handler
- * @rsc: reporting source code to handle
- */
-void s390_unregister_crw_handler(int rsc)
-{
-	if ((rsc < 0) || (rsc >= NR_RSCS))
-		return;
-	xchg(&crw_handlers[rsc], NULL);
-	synchronize_sched();
-}
-
-/*
- * Retrieve CRWs and call function to handle event.
- */
-static int s390_collect_crw_info(void *param)
-{
-	struct crw crw[2];
-	int ccode;
-	struct semaphore *sem;
-	unsigned int chain;
-	int ignore;
-
-	sem = (struct semaphore *)param;
-repeat:
-	ignore = down_interruptible(sem);
-	chain = 0;
-	while (1) {
-		if (unlikely(chain > 1)) {
-			struct crw tmp_crw;
-
-			printk(KERN_WARNING"%s: Code does not support more "
-			       "than two chained crws; please report to "
-			       "linux390@de.ibm.com!\n", __func__);
-			ccode = stcrw(&tmp_crw);
-			printk(KERN_WARNING"%s: crw reports slct=%d, oflw=%d, "
-			       "chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
-			       __func__, tmp_crw.slct, tmp_crw.oflw,
-			       tmp_crw.chn, tmp_crw.rsc, tmp_crw.anc,
-			       tmp_crw.erc, tmp_crw.rsid);
-			printk(KERN_WARNING"%s: This was crw number %x in the "
-			       "chain\n", __func__, chain);
-			if (ccode != 0)
-				break;
-			chain = tmp_crw.chn ? chain + 1 : 0;
-			continue;
-		}
-		ccode = stcrw(&crw[chain]);
-		if (ccode != 0)
-			break;
-		printk(KERN_DEBUG "crw_info : CRW reports slct=%d, oflw=%d, "
-		       "chn=%d, rsc=%X, anc=%d, erc=%X, rsid=%X\n",
-		       crw[chain].slct, crw[chain].oflw, crw[chain].chn,
-		       crw[chain].rsc, crw[chain].anc, crw[chain].erc,
-		       crw[chain].rsid);
-		/* Check for overflows. */
-		if (crw[chain].oflw) {
-			int i;
-
-			pr_debug("%s: crw overflow detected!\n", __func__);
-			for (i = 0; i < NR_RSCS; i++) {
-				if (crw_handlers[i])
-					crw_handlers[i](NULL, NULL, 1);
-			}
-			chain = 0;
-			continue;
-		}
-		if (crw[0].chn && !chain) {
-			chain++;
-			continue;
-		}
-		if (crw_handlers[crw[chain].rsc])
-			crw_handlers[crw[chain].rsc](&crw[0],
-						     chain ? &crw[1] : NULL,
-						     0);
-		/* chain is always 0 or 1 here. */
-		chain = crw[chain].chn ? chain + 1 : 0;
-	}
-	goto repeat;
-	return 0;
-}
-
-struct mcck_struct {
-	int kill_task;
-	int channel_report;
-	int warning;
-	unsigned long long mcck_code;
-};
-
-static DEFINE_PER_CPU(struct mcck_struct, cpu_mcck);
-
-/*
- * Main machine check handler function. Will be called with interrupts enabled
- * or disabled and machine checks enabled or disabled.
- */
-void
-s390_handle_mcck(void)
-{
-	unsigned long flags;
-	struct mcck_struct mcck;
-
-	/*
-	 * Disable machine checks and get the current state of accumulated
-	 * machine checks. Afterwards delete the old state and enable machine
-	 * checks again.
-	 */
-	local_irq_save(flags);
-	local_mcck_disable();
-	mcck = __get_cpu_var(cpu_mcck);
-	memset(&__get_cpu_var(cpu_mcck), 0, sizeof(struct mcck_struct));
-	clear_thread_flag(TIF_MCCK_PENDING);
-	local_mcck_enable();
-	local_irq_restore(flags);
-
-	if (mcck.channel_report)
-		up(&m_sem);
-
-#ifdef CONFIG_MACHCHK_WARNING
-/*
- * The warning may remain for a prolonged period on the bare iron.
- * (actually till the machine is powered off, or until the problem is gone)
- * So we just stop listening for the WARNING MCH and prevent continuously
- * being interrupted.  One caveat is however, that we must do this per
- * processor and cannot use the smp version of ctl_clear_bit().
- * On VM we only get one interrupt per virtally presented machinecheck.
- * Though one suffices, we may get one interrupt per (virtual) processor.
- */
-	if (mcck.warning) {	/* WARNING pending ? */
-		static int mchchk_wng_posted = 0;
-		/*
-		 * Use single machine clear, as we cannot handle smp right now
-		 */
-		__ctl_clear_bit(14, 24);	/* Disable WARNING MCH */
-		if (xchg(&mchchk_wng_posted, 1) == 0)
-			kill_cad_pid(SIGPWR, 1);
-	}
-#endif
-
-	if (mcck.kill_task) {
-		local_irq_enable();
-		printk(KERN_EMERG "mcck: Terminating task because of machine "
-		       "malfunction (code 0x%016llx).\n", mcck.mcck_code);
-		printk(KERN_EMERG "mcck: task: %s, pid: %d.\n",
-		       current->comm, current->pid);
-		do_exit(SIGSEGV);
-	}
-}
-EXPORT_SYMBOL_GPL(s390_handle_mcck);
-
-/*
- * returns 0 if all registers could be validated
- * returns 1 otherwise
- */
-static int
-s390_revalidate_registers(struct mci *mci)
-{
-	int kill_task;
-	u64 tmpclock;
-	u64 zero;
-	void *fpt_save_area, *fpt_creg_save_area;
-
-	kill_task = 0;
-	zero = 0;
-	/* General purpose registers */
-	if (!mci->gr)
-		/*
-		 * General purpose registers couldn't be restored and have
-		 * unknown contents. Process needs to be terminated.
-		 */
-		kill_task = 1;
-
-	/* Revalidate floating point registers */
-	if (!mci->fp)
-		/*
-		 * Floating point registers can't be restored and
-		 * therefore the process needs to be terminated.
-		 */
-		kill_task = 1;
-
-#ifndef CONFIG_64BIT
-	asm volatile(
-		"	ld	0,0(%0)\n"
-		"	ld	2,8(%0)\n"
-		"	ld	4,16(%0)\n"
-		"	ld	6,24(%0)"
-		: : "a" (&S390_lowcore.floating_pt_save_area));
-#endif
-
-	if (MACHINE_HAS_IEEE) {
-#ifdef CONFIG_64BIT
-		fpt_save_area = &S390_lowcore.floating_pt_save_area;
-		fpt_creg_save_area = &S390_lowcore.fpt_creg_save_area;
-#else
-		fpt_save_area = (void *) S390_lowcore.extended_save_area_addr;
-		fpt_creg_save_area = fpt_save_area+128;
-#endif
-		/* Floating point control register */
-		if (!mci->fc) {
-			/*
-			 * Floating point control register can't be restored.
-			 * Task will be terminated.
-			 */
-			asm volatile("lfpc 0(%0)" : : "a" (&zero), "m" (zero));
-			kill_task = 1;
-
-		} else
-			asm volatile("lfpc 0(%0)" : : "a" (fpt_creg_save_area));
-
-		asm volatile(
-			"	ld	0,0(%0)\n"
-			"	ld	1,8(%0)\n"
-			"	ld	2,16(%0)\n"
-			"	ld	3,24(%0)\n"
-			"	ld	4,32(%0)\n"
-			"	ld	5,40(%0)\n"
-			"	ld	6,48(%0)\n"
-			"	ld	7,56(%0)\n"
-			"	ld	8,64(%0)\n"
-			"	ld	9,72(%0)\n"
-			"	ld	10,80(%0)\n"
-			"	ld	11,88(%0)\n"
-			"	ld	12,96(%0)\n"
-			"	ld	13,104(%0)\n"
-			"	ld	14,112(%0)\n"
-			"	ld	15,120(%0)\n"
-			: : "a" (fpt_save_area));
-	}
-
-	/* Revalidate access registers */
-	asm volatile(
-		"	lam	0,15,0(%0)"
-		: : "a" (&S390_lowcore.access_regs_save_area));
-	if (!mci->ar)
-		/*
-		 * Access registers have unknown contents.
-		 * Terminating task.
-		 */
-		kill_task = 1;
-
-	/* Revalidate control registers */
-	if (!mci->cr)
-		/*
-		 * Control registers have unknown contents.
-		 * Can't recover and therefore stopping machine.
-		 */
-		s390_handle_damage("invalid control registers.");
-	else
-#ifdef CONFIG_64BIT
-		asm volatile(
-			"	lctlg	0,15,0(%0)"
-			: : "a" (&S390_lowcore.cregs_save_area));
-#else
-		asm volatile(
-			"	lctl	0,15,0(%0)"
-			: : "a" (&S390_lowcore.cregs_save_area));
-#endif
-
-	/*
-	 * We don't even try to revalidate the TOD register, since we simply
-	 * can't write something sensible into that register.
-	 */
-
-#ifdef CONFIG_64BIT
-	/*
-	 * See if we can revalidate the TOD programmable register with its
-	 * old contents (should be zero) otherwise set it to zero.
-	 */
-	if (!mci->pr)
-		asm volatile(
-			"	sr	0,0\n"
-			"	sckpf"
-			: : : "0", "cc");
-	else
-		asm volatile(
-			"	l	0,0(%0)\n"
-			"	sckpf"
-			: : "a" (&S390_lowcore.tod_progreg_save_area)
-			: "0", "cc");
-#endif
-
-	/* Revalidate clock comparator register */
-	asm volatile(
-		"	stck	0(%1)\n"
-		"	sckc	0(%1)"
-		: "=m" (tmpclock) : "a" (&(tmpclock)) : "cc", "memory");
-
-	/* Check if old PSW is valid */
-	if (!mci->wp)
-		/*
-		 * Can't tell if we come from user or kernel mode
-		 * -> stopping machine.
-		 */
-		s390_handle_damage("old psw invalid.");
-
-	if (!mci->ms || !mci->pm || !mci->ia)
-		kill_task = 1;
-
-	return kill_task;
-}
-
-#define MAX_IPD_COUNT	29
-#define MAX_IPD_TIME	(5 * 60 * USEC_PER_SEC) /* 5 minutes */
-
-/*
- * machine check handler.
- */
-void notrace s390_do_machine_check(struct pt_regs *regs)
-{
-	static DEFINE_SPINLOCK(ipd_lock);
-	static unsigned long long last_ipd;
-	static int ipd_count;
-	unsigned long long tmp;
-	struct mci *mci;
-	struct mcck_struct *mcck;
-	int umode;
-
-	lockdep_off();
-
-	s390_idle_check();
-
-	mci = (struct mci *) &S390_lowcore.mcck_interruption_code;
-	mcck = &__get_cpu_var(cpu_mcck);
-	umode = user_mode(regs);
-
-	if (mci->sd)
-		/* System damage -> stopping machine */
-		s390_handle_damage("received system damage machine check.");
-
-	if (mci->pd) {
-		if (mci->b) {
-			/* Processing backup -> verify if we can survive this */
-			u64 z_mcic, o_mcic, t_mcic;
-#ifdef CONFIG_64BIT
-			z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<29);
-			o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 |
-				  1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 |
-				  1ULL<<30 | 1ULL<<21 | 1ULL<<20 | 1ULL<<17 |
-				  1ULL<<16);
-#else
-			z_mcic = (1ULL<<63 | 1ULL<<59 | 1ULL<<57 | 1ULL<<50 |
-				  1ULL<<29);
-			o_mcic = (1ULL<<43 | 1ULL<<42 | 1ULL<<41 | 1ULL<<40 |
-				  1ULL<<36 | 1ULL<<35 | 1ULL<<34 | 1ULL<<32 |
-				  1ULL<<30 | 1ULL<<20 | 1ULL<<17 | 1ULL<<16);
-#endif
-			t_mcic = *(u64 *)mci;
-
-			if (((t_mcic & z_mcic) != 0) ||
-			    ((t_mcic & o_mcic) != o_mcic)) {
-				s390_handle_damage("processing backup machine "
-						   "check with damage.");
-			}
-
-			/*
-			 * Nullifying exigent condition, therefore we might
-			 * retry this instruction.
-			 */
-
-			spin_lock(&ipd_lock);
-
-			tmp = get_clock();
-
-			if (((tmp - last_ipd) >> 12) < MAX_IPD_TIME)
-				ipd_count++;
-			else
-				ipd_count = 1;
-
-			last_ipd = tmp;
-
-			if (ipd_count == MAX_IPD_COUNT)
-				s390_handle_damage("too many ipd retries.");
-
-			spin_unlock(&ipd_lock);
-		}
-		else {
-			/* Processing damage -> stopping machine */
-			s390_handle_damage("received instruction processing "
-					   "damage machine check.");
-		}
-	}
-	if (s390_revalidate_registers(mci)) {
-		if (umode) {
-			/*
-			 * Couldn't restore all register contents while in
-			 * user mode -> mark task for termination.
-			 */
-			mcck->kill_task = 1;
-			mcck->mcck_code = *(unsigned long long *) mci;
-			set_thread_flag(TIF_MCCK_PENDING);
-		}
-		else
-			/*
-			 * Couldn't restore all register contents while in
-			 * kernel mode -> stopping machine.
-			 */
-			s390_handle_damage("unable to revalidate registers.");
-	}
-
-	if (mci->cd) {
-		/* Timing facility damage */
-		s390_handle_damage("TOD clock damaged");
-	}
-
-	if (mci->ed && mci->ec) {
-		/* External damage */
-		if (S390_lowcore.external_damage_code & (1U << ED_ETR_SYNC))
-			etr_sync_check();
-		if (S390_lowcore.external_damage_code & (1U << ED_ETR_SWITCH))
-			etr_switch_to_local();
-		if (S390_lowcore.external_damage_code & (1U << ED_STP_SYNC))
-			stp_sync_check();
-		if (S390_lowcore.external_damage_code & (1U << ED_STP_ISLAND))
-			stp_island_check();
-	}
-
-	if (mci->se)
-		/* Storage error uncorrected */
-		s390_handle_damage("received storage error uncorrected "
-				   "machine check.");
-
-	if (mci->ke)
-		/* Storage key-error uncorrected */
-		s390_handle_damage("received storage key-error uncorrected "
-				   "machine check.");
-
-	if (mci->ds && mci->fa)
-		/* Storage degradation */
-		s390_handle_damage("received storage degradation machine "
-				   "check.");
-
-	if (mci->cp) {
-		/* Channel report word pending */
-		mcck->channel_report = 1;
-		set_thread_flag(TIF_MCCK_PENDING);
-	}
-
-	if (mci->w) {
-		/* Warning pending */
-		mcck->warning = 1;
-		set_thread_flag(TIF_MCCK_PENDING);
-	}
-	lockdep_on();
-}
-
-/*
- * s390_init_machine_check
- *
- * initialize machine check handling
- */
-static int
-machine_check_init(void)
-{
-	init_MUTEX_LOCKED(&m_sem);
-	ctl_set_bit(14, 25);	/* enable external damage MCH */
-	ctl_set_bit(14, 27);    /* enable system recovery MCH */
-#ifdef CONFIG_MACHCHK_WARNING
-	ctl_set_bit(14, 24);	/* enable warning MCH */
-#endif
-	return 0;
-}
-
-/*
- * Initialize the machine check handler really early to be able to
- * catch all machine checks that happen during boot
- */
-arch_initcall(machine_check_init);
-
-/*
- * Machine checks for the channel subsystem must be enabled
- * after the channel subsystem is initialized
- */
-static int __init
-machine_check_crw_init (void)
-{
-	struct task_struct *task;
-
-	task = kthread_run(s390_collect_crw_info, &m_sem, "kmcheck");
-	if (IS_ERR(task))
-		return PTR_ERR(task);
-	ctl_set_bit(14, 28);	/* enable channel report MCH */
-	return 0;
-}
-
-device_initcall (machine_check_crw_init);
diff --git a/drivers/s390/s390mach.h b/drivers/s390/s390mach.h
deleted file mode 100644
index d39f8b697d27..000000000000
--- a/drivers/s390/s390mach.h
+++ /dev/null
@@ -1,122 +0,0 @@
-/*
- *  drivers/s390/s390mach.h
- *   S/390 data definitions for machine check processing
- *
- *  S390 version
- *    Copyright (C) 2000 IBM Deutschland Entwicklung GmbH, IBM Corporation
- *    Author(s): Ingo Adlung (adlung@de.ibm.com)
- */
-
-#ifndef __s390mach_h
-#define __s390mach_h
-
-#include <asm/types.h>
-
-struct mci {
-	__u32   sd              :  1; /* 00 system damage */
-	__u32   pd              :  1; /* 01 instruction-processing damage */
-	__u32   sr              :  1; /* 02 system recovery */
-	__u32   to_be_defined_1 :  1; /* 03 */
-	__u32   cd              :  1; /* 04 timing-facility damage */
-	__u32   ed              :  1; /* 05 external damage */
-	__u32   to_be_defined_2 :  1; /* 06 */
-	__u32   dg              :  1; /* 07 degradation */
-	__u32   w               :  1; /* 08 warning pending */
-	__u32   cp              :  1; /* 09 channel-report pending */
-	__u32   sp              :  1; /* 10 service-processor damage */
-	__u32   ck              :  1; /* 11 channel-subsystem damage */
-	__u32   to_be_defined_3 :  2; /* 12-13 */
-	__u32   b               :  1; /* 14 backed up */
-	__u32   to_be_defined_4 :  1; /* 15 */
-	__u32   se              :  1; /* 16 storage error uncorrected */
-	__u32   sc              :  1; /* 17 storage error corrected */
-	__u32   ke              :  1; /* 18 storage-key error uncorrected */
-	__u32   ds              :  1; /* 19 storage degradation */
-	__u32   wp              :  1; /* 20 psw mwp validity */
-	__u32   ms              :  1; /* 21 psw mask and key validity */
-	__u32   pm              :  1; /* 22 psw program mask and cc validity */
-	__u32   ia              :  1; /* 23 psw instruction address validity */
-	__u32   fa              :  1; /* 24 failing storage address validity */
-	__u32   to_be_defined_5 :  1; /* 25 */
-	__u32   ec              :  1; /* 26 external damage code validity */
-	__u32   fp              :  1; /* 27 floating point register validity */
-	__u32   gr              :  1; /* 28 general register validity */
-	__u32   cr              :  1; /* 29 control register validity */
-	__u32   to_be_defined_6 :  1; /* 30 */
-	__u32   st              :  1; /* 31 storage logical validity */
-	__u32   ie              :  1; /* 32 indirect storage error */
-	__u32   ar              :  1; /* 33 access register validity */
-	__u32   da              :  1; /* 34 delayed access exception */
-	__u32   to_be_defined_7 :  7; /* 35-41 */
-	__u32   pr              :  1; /* 42 tod programmable register validity */
-	__u32   fc              :  1; /* 43 fp control register validity */
-	__u32   ap              :  1; /* 44 ancillary report */
-	__u32   to_be_defined_8 :  1; /* 45 */
-	__u32   ct              :  1; /* 46 cpu timer validity */
-	__u32   cc              :  1; /* 47 clock comparator validity */
-	__u32	to_be_defined_9 : 16; /* 47-63 */
-};
-
-/*
- * Channel Report Word
- */
-struct crw {
-	__u32 res1    :  1;   /* reserved zero */
-	__u32 slct    :  1;   /* solicited */
-	__u32 oflw    :  1;   /* overflow */
-	__u32 chn     :  1;   /* chained */
-	__u32 rsc     :  4;   /* reporting source code */
-	__u32 anc     :  1;   /* ancillary report */
-	__u32 res2    :  1;   /* reserved zero */
-	__u32 erc     :  6;   /* error-recovery code */
-	__u32 rsid    : 16;   /* reporting-source ID */
-} __attribute__ ((packed));
-
-typedef void (*crw_handler_t)(struct crw *, struct crw *, int);
-
-extern int s390_register_crw_handler(int rsc, crw_handler_t handler);
-extern void s390_unregister_crw_handler(int rsc);
-
-#define NR_RSCS 16
-
-#define CRW_RSC_MONITOR  0x2  /* monitoring facility */
-#define CRW_RSC_SCH      0x3  /* subchannel */
-#define CRW_RSC_CPATH    0x4  /* channel path */
-#define CRW_RSC_CONFIG   0x9  /* configuration-alert facility */
-#define CRW_RSC_CSS      0xB  /* channel subsystem */
-
-#define CRW_ERC_EVENT    0x00 /* event information pending */
-#define CRW_ERC_AVAIL    0x01 /* available */
-#define CRW_ERC_INIT     0x02 /* initialized */
-#define CRW_ERC_TERROR   0x03 /* temporary error */
-#define CRW_ERC_IPARM    0x04 /* installed parm initialized */
-#define CRW_ERC_TERM     0x05 /* terminal */
-#define CRW_ERC_PERRN    0x06 /* perm. error, fac. not init */
-#define CRW_ERC_PERRI    0x07 /* perm. error, facility init */
-#define CRW_ERC_PMOD     0x08 /* installed parameters modified */
-
-static inline int stcrw(struct crw *pcrw )
-{
-	int ccode;
-
-	__asm__ __volatile__(
-		"stcrw 0(%2)\n\t"
-		"ipm %0\n\t"
-		"srl %0,28\n\t"
-		: "=d" (ccode), "=m" (*pcrw)
-		: "a" (pcrw)
-		: "cc" );
-	return ccode;
-}
-
-#define ED_ETR_SYNC	12	/* External damage ETR sync check */
-#define ED_ETR_SWITCH	13	/* External damage ETR switch to local */
-
-#define ED_STP_SYNC	7	/* External damage STP sync check */
-#define ED_STP_ISLAND	6	/* External damage STP island check */
-
-struct pt_regs;
-
-void s390_handle_mcck(void);
-void s390_do_machine_check(struct pt_regs *regs);
-#endif /* __s390mach */
-- 
cgit v1.2.3-55-g7522


From 7b886416dfd76df9dd7304868556b1d82cf38890 Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:02 +0100
Subject: [S390] Remove CONFIG_MACHCHK_WARNING.

Everybody enables it so there is no point for an extra config option.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig      |  7 -------
 arch/s390/kernel/nmi.c | 29 +++++++++++------------------
 2 files changed, 11 insertions(+), 25 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 3f470604a89a..0a9463bea758 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -343,13 +343,6 @@ source "mm/Kconfig"
 
 comment "I/O subsystem configuration"
 
-config MACHCHK_WARNING
-	bool "Process warning machine checks"
-	help
-	  Select this option if you want the machine check handler on IBM S/390 or
-	  zSeries to process warning machine checks (e.g. on power failures).
-	  If unsure, say "Y".
-
 config QDIO
 	tristate "QDIO support"
 	---help---
diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c
index 8d50eaf76a60..4bfdc421d7e9 100644
--- a/arch/s390/kernel/nmi.c
+++ b/arch/s390/kernel/nmi.c
@@ -59,28 +59,23 @@ void s390_handle_mcck(void)
 
 	if (mcck.channel_report)
 		crw_handle_channel_report();
-
-#ifdef CONFIG_MACHCHK_WARNING
-/*
- * The warning may remain for a prolonged period on the bare iron.
- * (actually till the machine is powered off, or until the problem is gone)
- * So we just stop listening for the WARNING MCH and prevent continuously
- * being interrupted.  One caveat is however, that we must do this per
- * processor and cannot use the smp version of ctl_clear_bit().
- * On VM we only get one interrupt per virtally presented machinecheck.
- * Though one suffices, we may get one interrupt per (virtual) processor.
- */
+	/*
+	 * A warning may remain for a prolonged period on the bare iron.
+	 * (actually until the machine is powered off, or the problem is gone)
+	 * So we just stop listening for the WARNING MCH and avoid continuously
+	 * being interrupted.  One caveat is however, that we must do this per
+	 * processor and cannot use the smp version of ctl_clear_bit().
+	 * On VM we only get one interrupt per virtally presented machinecheck.
+	 * Though one suffices, we may get one interrupt per (virtual) cpu.
+	 */
 	if (mcck.warning) {	/* WARNING pending ? */
 		static int mchchk_wng_posted = 0;
-		/*
-		 * Use single machine clear, as we cannot handle smp right now
-		 */
+
+		/* Use single cpu clear, as we cannot handle smp here. */
 		__ctl_clear_bit(14, 24);	/* Disable WARNING MCH */
 		if (xchg(&mchchk_wng_posted, 1) == 0)
 			kill_cad_pid(SIGPWR, 1);
 	}
-#endif
-
 	if (mcck.kill_task) {
 		local_irq_enable();
 		printk(KERN_EMERG "mcck: Terminating task because of machine "
@@ -375,9 +370,7 @@ static int __init machine_check_init(void)
 {
 	ctl_set_bit(14, 25);	/* enable external damage MCH */
 	ctl_set_bit(14, 27);	/* enable system recovery MCH */
-#ifdef CONFIG_MACHCHK_WARNING
 	ctl_set_bit(14, 24);	/* enable warning MCH */
-#endif
 	return 0;
 }
 arch_initcall(machine_check_init);
-- 
cgit v1.2.3-55-g7522


From e3dd9c2da674993edb5b52acb56a5d954415639b Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:03 +0100
Subject: [S390] convert bitmap definitions to C

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/bitops.h |  2 +-
 arch/s390/kernel/bitmap.S      | 56 ------------------------------------------
 arch/s390/kernel/bitmap.c      | 49 ++++++++++++++++++++++++++++++++++++
 3 files changed, 50 insertions(+), 57 deletions(-)
 delete mode 100644 arch/s390/kernel/bitmap.S
 create mode 100644 arch/s390/kernel/bitmap.c

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/bitops.h b/arch/s390/include/asm/bitops.h
index 8e9243ae0c19..334964a0fb14 100644
--- a/arch/s390/include/asm/bitops.h
+++ b/arch/s390/include/asm/bitops.h
@@ -57,7 +57,7 @@
  * with operation of the form "set_bit(bitnr, flags)".
  */
 
-/* bitmap tables from arch/S390/kernel/bitmap.S */
+/* bitmap tables from arch/s390/kernel/bitmap.c */
 extern const char _oi_bitmap[];
 extern const char _ni_bitmap[];
 extern const char _zb_findmap[];
diff --git a/arch/s390/kernel/bitmap.S b/arch/s390/kernel/bitmap.S
deleted file mode 100644
index dfb41f946e23..000000000000
--- a/arch/s390/kernel/bitmap.S
+++ /dev/null
@@ -1,56 +0,0 @@
-/*
- *  arch/s390/kernel/bitmap.S
- *    Bitmaps for set_bit, clear_bit, test_and_set_bit, ...
- *    See include/asm-s390/{bitops.h|posix_types.h} for details
- *
- *  S390 version
- *    Copyright (C) 1999 IBM Deutschland Entwicklung GmbH, IBM Corporation
- *    Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com),
- */
-
-         .globl _oi_bitmap
-_oi_bitmap:
-         .byte  0x01,0x02,0x04,0x08,0x10,0x20,0x40,0x80
-
-         .globl _ni_bitmap
-_ni_bitmap:
-         .byte  0xFE,0xFD,0xFB,0xF7,0xEF,0xDF,0xBF,0x7F
-
-         .globl _zb_findmap
-_zb_findmap:
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4 
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,6
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,7
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,6
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4
-         .byte  0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,8
-
-         .globl _sb_findmap
-_sb_findmap:
-         .byte  8,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  6,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  7,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  6,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-         .byte  4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0
-
diff --git a/arch/s390/kernel/bitmap.c b/arch/s390/kernel/bitmap.c
new file mode 100644
index 000000000000..54e362b4267d
--- /dev/null
+++ b/arch/s390/kernel/bitmap.c
@@ -0,0 +1,49 @@
+/*
+ *    Bitmaps for set_bit, clear_bit, test_and_set_bit, ...
+ *    See include/asm/{bitops.h|posix_types.h} for details
+ *
+ *    Copyright IBM Corp. 1999,2009
+ *    Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>,
+ */
+
+#include <linux/bitops.h>
+
+const char _oi_bitmap[] = { 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80 };
+
+const char _ni_bitmap[] = { 0xfe, 0xfd, 0xfb, 0xf7, 0xef, 0xdf, 0xbf, 0x7f };
+
+const char _zb_findmap[] = {
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,6,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,7,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,6,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
+	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,8 };
+
+const char _sb_findmap[] = {
+	8,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	6,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	7,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	6,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
+	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0 };
-- 
cgit v1.2.3-55-g7522


From 1485c5c88483d200c9c4c71ed7e8eef1a1e317a1 Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:04 +0100
Subject: [S390] move EXPORT_SYMBOLs to definitions

Move all EXPORT_SYMBOLs to their corresponding definitions.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/bitmap.c     |  5 +++++
 arch/s390/kernel/process.c    |  2 ++
 arch/s390/kernel/s390_ksyms.c | 44 -------------------------------------------
 arch/s390/kernel/setup.c      |  8 ++++++++
 arch/s390/lib/delay.c         |  2 ++
 arch/s390/mm/init.c           |  2 ++
 6 files changed, 19 insertions(+), 44 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/bitmap.c b/arch/s390/kernel/bitmap.c
index 54e362b4267d..3ae4757b006a 100644
--- a/arch/s390/kernel/bitmap.c
+++ b/arch/s390/kernel/bitmap.c
@@ -7,10 +7,13 @@
  */
 
 #include <linux/bitops.h>
+#include <linux/module.h>
 
 const char _oi_bitmap[] = { 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x40, 0x80 };
+EXPORT_SYMBOL(_oi_bitmap);
 
 const char _ni_bitmap[] = { 0xfe, 0xfd, 0xfb, 0xf7, 0xef, 0xdf, 0xbf, 0x7f };
+EXPORT_SYMBOL(_ni_bitmap);
 
 const char _zb_findmap[] = {
 	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
@@ -29,6 +32,7 @@ const char _zb_findmap[] = {
 	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,5,
 	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,4,
 	0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,8 };
+EXPORT_SYMBOL(_zb_findmap);
 
 const char _sb_findmap[] = {
 	8,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
@@ -47,3 +51,4 @@ const char _sb_findmap[] = {
 	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
 	5,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0,
 	4,0,1,0,2,0,1,0,3,0,1,0,2,0,1,0 };
+EXPORT_SYMBOL(_sb_findmap);
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index bd616fb31e75..b48e961a38f6 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -141,6 +141,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
 	return do_fork(flags | CLONE_VM | CLONE_UNTRACED,
 		       0, &regs, 0, NULL, NULL);
 }
+EXPORT_SYMBOL(kernel_thread);
 
 /*
  * Free current thread data structures etc..
@@ -318,6 +319,7 @@ int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs)
 #endif /* CONFIG_64BIT */
 	return 1;
 }
+EXPORT_SYMBOL(dump_fpu);
 
 unsigned long get_wchan(struct task_struct *p)
 {
diff --git a/arch/s390/kernel/s390_ksyms.c b/arch/s390/kernel/s390_ksyms.c
index 46b90cb03707..656fcbb9bd83 100644
--- a/arch/s390/kernel/s390_ksyms.c
+++ b/arch/s390/kernel/s390_ksyms.c
@@ -1,49 +1,5 @@
-/*
- *  arch/s390/kernel/s390_ksyms.c
- *
- *  S390 version
- */
-#include <linux/highuid.h>
 #include <linux/module.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/syscalls.h>
-#include <linux/interrupt.h>
-#include <asm/checksum.h>
-#include <asm/cpcmd.h>
-#include <asm/delay.h>
-#include <asm/pgalloc.h>
-#include <asm/setup.h>
 #include <asm/ftrace.h>
-#ifdef CONFIG_IP_MULTICAST
-#include <net/arp.h>
-#endif
-
-/*
- * memory management
- */
-EXPORT_SYMBOL(_oi_bitmap);
-EXPORT_SYMBOL(_ni_bitmap);
-EXPORT_SYMBOL(_zb_findmap);
-EXPORT_SYMBOL(_sb_findmap);
-
-/*
- * binfmt_elf loader 
- */
-extern int dump_fpu (struct pt_regs * regs, s390_fp_regs *fpregs);
-EXPORT_SYMBOL(dump_fpu);
-EXPORT_SYMBOL(empty_zero_page);
-
-/*
- * misc.
- */
-EXPORT_SYMBOL(machine_flags);
-EXPORT_SYMBOL(__udelay);
-EXPORT_SYMBOL(kernel_thread);
-EXPORT_SYMBOL(csum_fold);
-EXPORT_SYMBOL(console_mode);
-EXPORT_SYMBOL(console_devno);
-EXPORT_SYMBOL(console_irq);
 
 #ifdef CONFIG_FUNCTION_TRACER
 EXPORT_SYMBOL(_mcount);
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 8fdf08379ce9..0b8ad7b4f8f3 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -74,9 +74,17 @@ EXPORT_SYMBOL(uaccess);
  * Machine setup..
  */
 unsigned int console_mode = 0;
+EXPORT_SYMBOL(console_mode);
+
 unsigned int console_devno = -1;
+EXPORT_SYMBOL(console_devno);
+
 unsigned int console_irq = -1;
+EXPORT_SYMBOL(console_irq);
+
 unsigned long machine_flags;
+EXPORT_SYMBOL(machine_flags);
+
 unsigned long elf_hwcap = 0;
 char elf_platform[ELF_PLATFORM_SIZE];
 
diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c
index 6ccb9fab055a..3f5f680726ed 100644
--- a/arch/s390/lib/delay.c
+++ b/arch/s390/lib/delay.c
@@ -9,6 +9,7 @@
 #include <linux/sched.h>
 #include <linux/delay.h>
 #include <linux/timex.h>
+#include <linux/module.h>
 #include <linux/irqflags.h>
 #include <linux/interrupt.h>
 
@@ -92,6 +93,7 @@ out:
 	local_irq_restore(flags);
 	preempt_enable();
 }
+EXPORT_SYMBOL(__udelay);
 
 /*
  * Simple udelay variant. To be used on startup and reboot
diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index f0258ca3b17e..c634dfbe92e9 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -40,7 +40,9 @@
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
 pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__((__aligned__(PAGE_SIZE)));
+
 char  empty_zero_page[PAGE_SIZE] __attribute__((__aligned__(PAGE_SIZE)));
+EXPORT_SYMBOL(empty_zero_page);
 
 /*
  * paging_init() sets up the page tables
-- 
cgit v1.2.3-55-g7522


From 8283cb43ab3e92a039d3486a0f6253df95a5fa55 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky
Date: Thu, 26 Mar 2009 15:24:21 +0100
Subject: [S390] clock sync mode flags

The clock sync mode flag CLOCK_SYNC_STP is not cleared when stp
is set offline. In this case the get_sync_clock() function returns
-EACCESS and the dasd driver will block all i/o until stp is enabled
again. In addition get_sync_clock can return -EACCESS if the clock is
not in sync instead of -EAGAIN.

Rework the stp/etr online handling to fix these problems.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/time.c | 71 ++++++++++++++++++++++++++++++-------------------
 1 file changed, 44 insertions(+), 27 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c
index fc468cae4460..f72d41068dc2 100644
--- a/arch/s390/kernel/time.c
+++ b/arch/s390/kernel/time.c
@@ -331,6 +331,7 @@ static unsigned long long adjust_time(unsigned long long old,
 }
 
 static DEFINE_PER_CPU(atomic_t, clock_sync_word);
+static DEFINE_MUTEX(clock_sync_mutex);
 static unsigned long clock_sync_flags;
 
 #define CLOCK_SYNC_HAS_ETR	0
@@ -394,6 +395,20 @@ static void enable_sync_clock(void)
 	atomic_set_mask(0x80000000, sw_ptr);
 }
 
+/*
+ * Function to check if the clock is in sync.
+ */
+static inline int check_sync_clock(void)
+{
+	atomic_t *sw_ptr;
+	int rc;
+
+	sw_ptr = &get_cpu_var(clock_sync_word);
+	rc = (atomic_read(sw_ptr) & 0x80000000U) != 0;
+	put_cpu_var(clock_sync_sync);
+	return rc;
+}
+
 /* Single threaded workqueue used for etr and stp sync events */
 static struct workqueue_struct *time_sync_wq;
 
@@ -485,6 +500,8 @@ static void etr_reset(void)
 	if (etr_setr(&etr_eacr) == 0) {
 		etr_tolec = get_clock();
 		set_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags);
+		if (etr_port0_online && etr_port1_online)
+			set_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
 	} else if (etr_port0_online || etr_port1_online) {
 		pr_warning("The real or virtual hardware system does "
 			   "not provide an ETR interface\n");
@@ -533,8 +550,7 @@ void etr_switch_to_local(void)
 {
 	if (!etr_eacr.sl)
 		return;
-	if (test_bit(CLOCK_SYNC_ETR, &clock_sync_flags))
-		disable_sync_clock(NULL);
+	disable_sync_clock(NULL);
 	set_bit(ETR_EVENT_SWITCH_LOCAL, &etr_events);
 	queue_work(time_sync_wq, &etr_work);
 }
@@ -549,8 +565,7 @@ void etr_sync_check(void)
 {
 	if (!etr_eacr.es)
 		return;
-	if (test_bit(CLOCK_SYNC_ETR, &clock_sync_flags))
-		disable_sync_clock(NULL);
+	disable_sync_clock(NULL);
 	set_bit(ETR_EVENT_SYNC_CHECK, &etr_events);
 	queue_work(time_sync_wq, &etr_work);
 }
@@ -914,7 +929,7 @@ static struct etr_eacr etr_handle_update(struct etr_aib *aib,
 	 * Do not try to get the alternate port aib if the clock
 	 * is not in sync yet.
 	 */
-	if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags) && !eacr.es)
+	if (!check_sync_clock())
 		return eacr;
 
 	/*
@@ -997,7 +1012,6 @@ static void etr_work_fn(struct work_struct *work)
 		on_each_cpu(disable_sync_clock, NULL, 1);
 		del_timer_sync(&etr_timer);
 		etr_update_eacr(eacr);
-		clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
 		goto out_unlock;
 	}
 
@@ -1071,18 +1085,13 @@ static void etr_work_fn(struct work_struct *work)
 		/* Both ports not usable. */
 		eacr.es = eacr.sl = 0;
 		sync_port = -1;
-		clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
 	}
 
-	if (!test_bit(CLOCK_SYNC_ETR, &clock_sync_flags))
-		eacr.es = 0;
-
 	/*
 	 * If the clock is in sync just update the eacr and return.
 	 * If there is no valid sync port wait for a port update.
 	 */
-	if (test_bit(CLOCK_SYNC_STP, &clock_sync_flags) ||
-	    eacr.es || sync_port < 0) {
+	if (check_sync_clock() || sync_port < 0) {
 		etr_update_eacr(eacr);
 		etr_set_tolec_timeout(now);
 		goto out_unlock;
@@ -1103,13 +1112,11 @@ static void etr_work_fn(struct work_struct *work)
 	 * and set up a timer to try again after 0.5 seconds
 	 */
 	etr_update_eacr(eacr);
-	set_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
 	if (now < etr_tolec + (1600000 << 12) ||
 	    etr_sync_clock_stop(&aib, sync_port) != 0) {
 		/* Sync failed. Try again in 1/2 second. */
 		eacr.es = 0;
 		etr_update_eacr(eacr);
-		clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
 		etr_set_sync_timeout();
 	} else
 		etr_set_tolec_timeout(now);
@@ -1191,19 +1198,30 @@ static ssize_t etr_online_store(struct sys_device *dev,
 		return -EINVAL;
 	if (!test_bit(CLOCK_SYNC_HAS_ETR, &clock_sync_flags))
 		return -EOPNOTSUPP;
+	mutex_lock(&clock_sync_mutex);
 	if (dev == &etr_port0_dev) {
 		if (etr_port0_online == value)
-			return count;	/* Nothing to do. */
+			goto out;	/* Nothing to do. */
 		etr_port0_online = value;
+		if (etr_port0_online && etr_port1_online)
+			set_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
+		else
+			clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
 		set_bit(ETR_EVENT_PORT0_CHANGE, &etr_events);
 		queue_work(time_sync_wq, &etr_work);
 	} else {
 		if (etr_port1_online == value)
-			return count;	/* Nothing to do. */
+			goto out;	/* Nothing to do. */
 		etr_port1_online = value;
+		if (etr_port0_online && etr_port1_online)
+			set_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
+		else
+			clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags);
 		set_bit(ETR_EVENT_PORT1_CHANGE, &etr_events);
 		queue_work(time_sync_wq, &etr_work);
 	}
+out:
+	mutex_unlock(&clock_sync_mutex);
 	return count;
 }
 
@@ -1471,8 +1489,6 @@ static void stp_timing_alert(struct stp_irq_parm *intparm)
  */
 void stp_sync_check(void)
 {
-	if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags))
-		return;
 	disable_sync_clock(NULL);
 	queue_work(time_sync_wq, &stp_work);
 }
@@ -1485,8 +1501,6 @@ void stp_sync_check(void)
  */
 void stp_island_check(void)
 {
-	if (!test_bit(CLOCK_SYNC_STP, &clock_sync_flags))
-		return;
 	disable_sync_clock(NULL);
 	queue_work(time_sync_wq, &stp_work);
 }
@@ -1513,10 +1527,6 @@ static int stp_sync_clock(void *data)
 
 	enable_sync_clock();
 
-	set_bit(CLOCK_SYNC_STP, &clock_sync_flags);
-	if (test_and_clear_bit(CLOCK_SYNC_ETR, &clock_sync_flags))
-		queue_work(time_sync_wq, &etr_work);
-
 	rc = 0;
 	if (stp_info.todoff[0] || stp_info.todoff[1] ||
 	    stp_info.todoff[2] || stp_info.todoff[3] ||
@@ -1535,9 +1545,6 @@ static int stp_sync_clock(void *data)
 	if (rc) {
 		disable_sync_clock(NULL);
 		stp_sync->in_sync = -EAGAIN;
-		clear_bit(CLOCK_SYNC_STP, &clock_sync_flags);
-		if (etr_port0_online || etr_port1_online)
-			queue_work(time_sync_wq, &etr_work);
 	} else
 		stp_sync->in_sync = 1;
 	xchg(&first, 0);
@@ -1569,6 +1576,10 @@ static void stp_work_fn(struct work_struct *work)
 	if (rc || stp_info.c == 0)
 		goto out_unlock;
 
+	/* Skip synchronization if the clock is already in sync. */
+	if (check_sync_clock())
+		goto out_unlock;
+
 	memset(&stp_sync, 0, sizeof(stp_sync));
 	get_online_cpus();
 	atomic_set(&stp_sync.cpus, num_online_cpus() - 1);
@@ -1684,8 +1695,14 @@ static ssize_t stp_online_store(struct sysdev_class *class,
 		return -EINVAL;
 	if (!test_bit(CLOCK_SYNC_HAS_STP, &clock_sync_flags))
 		return -EOPNOTSUPP;
+	mutex_lock(&clock_sync_mutex);
 	stp_online = value;
+	if (stp_online)
+		set_bit(CLOCK_SYNC_STP, &clock_sync_flags);
+	else
+		clear_bit(CLOCK_SYNC_STP, &clock_sync_flags);
 	queue_work(time_sync_wq, &stp_work);
+	mutex_unlock(&clock_sync_mutex);
 	return count;
 }
 
-- 
cgit v1.2.3-55-g7522


From 4b2a8b6043979f1e6cc8aaa4116c95b31ae942ec Mon Sep 17 00:00:00 2001
From: Gerald Schaefer
Date: Thu, 26 Mar 2009 15:24:22 +0100
Subject: [S390] kernel: Disable switch_amode by default

Disable switch_amode by default because pagetable walk on pre z9
hardware has negative performance impact.

Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/setup.c | 4 ----
 1 file changed, 4 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 0b8ad7b4f8f3..dd3c51736270 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -301,11 +301,7 @@ static int __init early_parse_mem(char *p)
 early_param("mem", early_parse_mem);
 
 #ifdef CONFIG_S390_SWITCH_AMODE
-#ifdef CONFIG_PGSTE
-unsigned int switch_amode = 1;
-#else
 unsigned int switch_amode = 0;
-#endif
 EXPORT_SYMBOL_GPL(switch_amode);
 
 static int set_amode_and_uaccess(unsigned long user_amode,
-- 
cgit v1.2.3-55-g7522


From d303b6fd858370c22d5c70c313669e3521a5f758 Mon Sep 17 00:00:00 2001
From: Jan Glauber
Date: Thu, 26 Mar 2009 15:24:31 +0100
Subject: [S390] qdio: report SIGA errors directly

Errors from SIGA instructions are stored in the per queue qdio_error
and reported back when the queue handler is called. That opens a race
when multiple error conditions occur simultanously.

Report SIGA errors immediately in the return value of do_QDIO so the
upper layer can react and SIGA errors no longer interfere with other
errors.

Move the SIGA error handling in qeth from the outbound handler to
qeth_flush_buffers.

Signed-off-by: Jan Glauber <jang@linux.vnet.ibm.com>
---
 arch/s390/include/asm/qdio.h      |  1 +
 drivers/s390/cio/qdio.h           |  1 -
 drivers/s390/cio/qdio_main.c      | 73 ++++++++++++++++-----------------------
 drivers/s390/cio/qdio_setup.c     |  1 -
 drivers/s390/net/qeth_core_main.c | 55 +++++++++--------------------
 5 files changed, 48 insertions(+), 83 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/qdio.h b/arch/s390/include/asm/qdio.h
index 27fc1746de15..402d6dcf0d26 100644
--- a/arch/s390/include/asm/qdio.h
+++ b/arch/s390/include/asm/qdio.h
@@ -314,6 +314,7 @@ typedef void qdio_handler_t(struct ccw_device *, unsigned int, int,
 			    int, int, unsigned long);
 
 /* qdio errors reported to the upper-layer program */
+#define QDIO_ERROR_SIGA_TARGET			0x02
 #define QDIO_ERROR_SIGA_ACCESS_EXCEPTION	0x10
 #define QDIO_ERROR_SIGA_BUSY			0x20
 #define QDIO_ERROR_ACTIVATE_CHECK_CONDITION	0x40
diff --git a/drivers/s390/cio/qdio.h b/drivers/s390/cio/qdio.h
index 57807f5ffe84..41171d741f38 100644
--- a/drivers/s390/cio/qdio.h
+++ b/drivers/s390/cio/qdio.h
@@ -247,7 +247,6 @@ struct qdio_q {
 
 	struct qdio_irq *irq_ptr;
 	struct tasklet_struct tasklet;
-	spinlock_t lock;
 
 	/* error condition during a data transfer */
 	unsigned int qdio_error;
diff --git a/drivers/s390/cio/qdio_main.c b/drivers/s390/cio/qdio_main.c
index 31b9318149ba..e53ac67e1e48 100644
--- a/drivers/s390/cio/qdio_main.c
+++ b/drivers/s390/cio/qdio_main.c
@@ -706,13 +706,13 @@ static inline int qdio_outbound_q_moved(struct qdio_q *q)
 		return 0;
 }
 
-static void qdio_kick_outbound_q(struct qdio_q *q)
+static int qdio_kick_outbound_q(struct qdio_q *q)
 {
 	unsigned int busy_bit;
 	int cc;
 
 	if (!need_siga_out(q))
-		return;
+		return 0;
 
 	DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-w:%1d", q->nr);
 	qdio_perf_stat_inc(&perf_stats.siga_out);
@@ -724,19 +724,16 @@ static void qdio_kick_outbound_q(struct qdio_q *q)
 	case 2:
 		if (busy_bit) {
 			DBF_ERROR("%4x cc2 REP:%1d", SCH_NO(q), q->nr);
-			q->qdio_error = cc | QDIO_ERROR_SIGA_BUSY;
-		} else {
-			DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-w cc2:%1d",
-				      q->nr);
-			q->qdio_error = cc;
-		}
+			cc |= QDIO_ERROR_SIGA_BUSY;
+		} else
+			DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "siga-w cc2:%1d", q->nr);
 		break;
 	case 1:
 	case 3:
 		DBF_ERROR("%4x SIGA-W:%1d", SCH_NO(q), cc);
-		q->qdio_error = cc;
 		break;
 	}
+	return cc;
 }
 
 static void qdio_kick_outbound_handler(struct qdio_q *q)
@@ -766,18 +763,12 @@ static void qdio_kick_outbound_handler(struct qdio_q *q)
 
 static void __qdio_outbound_processing(struct qdio_q *q)
 {
-	unsigned long flags;
-
 	qdio_perf_stat_inc(&perf_stats.tasklet_outbound);
-	spin_lock_irqsave(&q->lock, flags);
-
 	BUG_ON(atomic_read(&q->nr_buf_used) < 0);
 
 	if (qdio_outbound_q_moved(q))
 		qdio_kick_outbound_handler(q);
 
-	spin_unlock_irqrestore(&q->lock, flags);
-
 	if (queue_type(q) == QDIO_ZFCP_QFMT)
 		if (!pci_out_supported(q) && !qdio_outbound_q_done(q))
 			goto sched;
@@ -1457,10 +1448,10 @@ static inline int buf_in_between(int bufnr, int start, int count)
  * @bufnr: first buffer to process
  * @count: how many buffers are emptied
  */
-static void handle_inbound(struct qdio_q *q, unsigned int callflags,
-			   int bufnr, int count)
+static int handle_inbound(struct qdio_q *q, unsigned int callflags,
+			  int bufnr, int count)
 {
-	int used, cc, diff;
+	int used, diff;
 
 	if (!q->u.in.polling)
 		goto set;
@@ -1497,13 +1488,11 @@ set:
 
 	/* no need to signal as long as the adapter had free buffers */
 	if (used)
-		return;
+		return 0;
 
-	if (need_siga_in(q)) {
-		cc = qdio_siga_input(q);
-		if (cc)
-			q->qdio_error = cc;
-	}
+	if (need_siga_in(q))
+		return qdio_siga_input(q);
+	return 0;
 }
 
 /**
@@ -1513,11 +1502,11 @@ set:
  * @bufnr: first buffer to process
  * @count: how many buffers are filled
  */
-static void handle_outbound(struct qdio_q *q, unsigned int callflags,
-			    int bufnr, int count)
+static int handle_outbound(struct qdio_q *q, unsigned int callflags,
+			   int bufnr, int count)
 {
 	unsigned char state;
-	int used;
+	int used, rc = 0;
 
 	qdio_perf_stat_inc(&perf_stats.outbound_handler);
 
@@ -1532,27 +1521,26 @@ static void handle_outbound(struct qdio_q *q, unsigned int callflags,
 
 	if (queue_type(q) == QDIO_IQDIO_QFMT) {
 		if (multicast_outbound(q))
-			qdio_kick_outbound_q(q);
+			rc = qdio_kick_outbound_q(q);
 		else
 			if ((q->irq_ptr->ssqd_desc.mmwc > 1) &&
 			    (count > 1) &&
 			    (count <= q->irq_ptr->ssqd_desc.mmwc)) {
 				/* exploit enhanced SIGA */
 				q->u.out.use_enh_siga = 1;
-				qdio_kick_outbound_q(q);
+				rc = qdio_kick_outbound_q(q);
 			} else {
 				/*
 				* One siga-w per buffer required for unicast
 				* HiperSockets.
 				*/
 				q->u.out.use_enh_siga = 0;
-				while (count--)
-					qdio_kick_outbound_q(q);
+				while (count--) {
+					rc = qdio_kick_outbound_q(q);
+					if (rc)
+						goto out;
+				}
 			}
-
-		/* report CC=2 conditions synchronously */
-		if (q->qdio_error)
-			__qdio_outbound_processing(q);
 		goto out;
 	}
 
@@ -1564,13 +1552,14 @@ static void handle_outbound(struct qdio_q *q, unsigned int callflags,
 	/* try to fast requeue buffers */
 	get_buf_state(q, prev_buf(bufnr), &state, 0);
 	if (state != SLSB_CU_OUTPUT_PRIMED)
-		qdio_kick_outbound_q(q);
+		rc = qdio_kick_outbound_q(q);
 	else {
 		DBF_DEV_EVENT(DBF_INFO, q->irq_ptr, "fast-req");
 		qdio_perf_stat_inc(&perf_stats.fast_requeue);
 	}
 out:
 	tasklet_schedule(&q->tasklet);
+	return rc;
 }
 
 /**
@@ -1609,14 +1598,12 @@ int do_QDIO(struct ccw_device *cdev, unsigned int callflags,
 		return -EBUSY;
 
 	if (callflags & QDIO_FLAG_SYNC_INPUT)
-		handle_inbound(irq_ptr->input_qs[q_nr], callflags, bufnr,
-			       count);
+		return handle_inbound(irq_ptr->input_qs[q_nr],
+				      callflags, bufnr, count);
 	else if (callflags & QDIO_FLAG_SYNC_OUTPUT)
-		handle_outbound(irq_ptr->output_qs[q_nr], callflags, bufnr,
-				count);
-	else
-		return -EINVAL;
-	return 0;
+		return handle_outbound(irq_ptr->output_qs[q_nr],
+				       callflags, bufnr, count);
+	return -EINVAL;
 }
 EXPORT_SYMBOL_GPL(do_QDIO);
 
diff --git a/drivers/s390/cio/qdio_setup.c b/drivers/s390/cio/qdio_setup.c
index c08356b95bf5..18d54fc21ce9 100644
--- a/drivers/s390/cio/qdio_setup.c
+++ b/drivers/s390/cio/qdio_setup.c
@@ -117,7 +117,6 @@ static void setup_queues_misc(struct qdio_q *q, struct qdio_irq *irq_ptr,
 	q->mask = 1 << (31 - i);
 	q->nr = i;
 	q->handler = handler;
-	spin_lock_init(&q->lock);
 }
 
 static void setup_storage_lists(struct qdio_q *q, struct qdio_irq *irq_ptr,
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index d1b5bebea7fb..2489bcebf5ec 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -2693,40 +2693,21 @@ static int qeth_handle_send_error(struct qeth_card *card,
 		struct qeth_qdio_out_buffer *buffer, unsigned int qdio_err)
 {
 	int sbalf15 = buffer->buffer->element[15].flags & 0xff;
-	int cc = qdio_err & 3;
 
 	QETH_DBF_TEXT(TRACE, 6, "hdsnderr");
 	qeth_check_qdio_errors(buffer->buffer, qdio_err, "qouterr");
-	switch (cc) {
-	case 0:
-		if (qdio_err) {
-			QETH_DBF_TEXT(TRACE, 1, "lnkfail");
-			QETH_DBF_TEXT_(TRACE, 1, "%s", CARD_BUS_ID(card));
-			QETH_DBF_TEXT_(TRACE, 1, "%04x %02x",
-				       (u16)qdio_err, (u8)sbalf15);
-			return QETH_SEND_ERROR_LINK_FAILURE;
-		}
+
+	if (!qdio_err)
 		return QETH_SEND_ERROR_NONE;
-	case 2:
-		if (qdio_err & QDIO_ERROR_SIGA_BUSY) {
-			QETH_DBF_TEXT(TRACE, 1, "SIGAcc2B");
-			QETH_DBF_TEXT_(TRACE, 1, "%s", CARD_BUS_ID(card));
-			return QETH_SEND_ERROR_KICK_IT;
-		}
-		if ((sbalf15 >= 15) && (sbalf15 <= 31))
-			return QETH_SEND_ERROR_RETRY;
-		return QETH_SEND_ERROR_LINK_FAILURE;
-		/* look at qdio_error and sbalf 15 */
-	case 1:
-		QETH_DBF_TEXT(TRACE, 1, "SIGAcc1");
-		QETH_DBF_TEXT_(TRACE, 1, "%s", CARD_BUS_ID(card));
-		return QETH_SEND_ERROR_LINK_FAILURE;
-	case 3:
-	default:
-		QETH_DBF_TEXT(TRACE, 1, "SIGAcc3");
-		QETH_DBF_TEXT_(TRACE, 1, "%s", CARD_BUS_ID(card));
-		return QETH_SEND_ERROR_KICK_IT;
-	}
+
+	if ((sbalf15 >= 15) && (sbalf15 <= 31))
+		return QETH_SEND_ERROR_RETRY;
+
+	QETH_DBF_TEXT(TRACE, 1, "lnkfail");
+	QETH_DBF_TEXT_(TRACE, 1, "%s", CARD_BUS_ID(card));
+	QETH_DBF_TEXT_(TRACE, 1, "%04x %02x",
+		       (u16)qdio_err, (u8)sbalf15);
+	return QETH_SEND_ERROR_LINK_FAILURE;
 }
 
 /*
@@ -2862,10 +2843,14 @@ static void qeth_flush_buffers(struct qeth_qdio_out_q *queue, int index,
 			qeth_get_micros() -
 			queue->card->perf_stats.outbound_do_qdio_start_time;
 	if (rc) {
+		queue->card->stats.tx_errors += count;
+		/* ignore temporary SIGA errors without busy condition */
+		if (rc == QDIO_ERROR_SIGA_TARGET)
+			return;
 		QETH_DBF_TEXT(TRACE, 2, "flushbuf");
 		QETH_DBF_TEXT_(TRACE, 2, " err%d", rc);
 		QETH_DBF_TEXT_(TRACE, 2, "%s", CARD_DDEV_ID(queue->card));
-		queue->card->stats.tx_errors += count;
+
 		/* this must not happen under normal circumstances. if it
 		 * happens something is really wrong -> recover */
 		qeth_schedule_recovery(queue->card);
@@ -2940,13 +2925,7 @@ void qeth_qdio_output_handler(struct ccw_device *ccwdev,
 	}
 	for (i = first_element; i < (first_element + count); ++i) {
 		buffer = &queue->bufs[i % QDIO_MAX_BUFFERS_PER_Q];
-		/*we only handle the KICK_IT error by doing a recovery */
-		if (qeth_handle_send_error(card, buffer, qdio_error)
-				== QETH_SEND_ERROR_KICK_IT){
-			netif_stop_queue(card->dev);
-			qeth_schedule_recovery(card);
-			return;
-		}
+		qeth_handle_send_error(card, buffer, qdio_error);
 		qeth_clear_output_buffer(queue, buffer);
 	}
 	atomic_sub(count, &queue->used_buffers);
-- 
cgit v1.2.3-55-g7522


From 6faf250789dfd35146a71402f9be245be82bf01e Mon Sep 17 00:00:00 2001
From: Rusty Russell
Date: Thu, 26 Mar 2009 15:24:33 +0100
Subject: [S390] allow usage of string functions in linux/string.h

In introducing a trivial "strstarts()" function in linux/string.h, we
hit the following error on s390:

In file included from include/linux/bitmap.h:8,
                 from include/linux/cpumask.h:142,
                 from include/linux/smp.h:12,
                 from /home/rusty/devel/kernel/patches/linux-2.6/arch/s390/include/asm/spinlock.h:14,
                 from include/linux/spinlock.h:88,
                 from include/linux/seqlock.h:29,
                 from include/linux/time.h:8,
                 from include/linux/stat.h:60,
                 from include/linux/module.h:10,
                 from arch/s390/lib/string.c:13:
include/linux/string.h: In function 'strstarts':
include/linux/string.h:124: error: implicit declaration of function 'strlen'
include/linux/string.h:124: warning: incompatible implicit declaration of built-in function 'strlen'

Because when including asm/string.h from arch/s390/lib/string.c we
don't declare the string ops we are about to define, and
linux/string.h barfs.

The fix is to declare them in this IN_ARCH_STRING_C case, but in
general I wonder if there's a neater fix.

Reported-by: linux-next
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/string.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h
index d074673a6d9b..adf079170aa6 100644
--- a/arch/s390/include/asm/string.h
+++ b/arch/s390/include/asm/string.h
@@ -135,7 +135,13 @@ static inline size_t strnlen(const char * s, size_t n)
 		: "+a" (end), "+a" (tmp) : "d" (r0)  : "cc");
 	return end - s;
 }
-
+#else /* IN_ARCH_STRING_C */
+void *memchr(const void * s, int c, size_t n);
+void *memscan(void *s, int c, size_t n);
+char *strcat(char *dst, const char *src);
+char *strcpy(char *dst, const char *src);
+size_t strlen(const char *s);
+size_t strnlen(const char * s, size_t n);
 #endif /* !IN_ARCH_STRING_C */
 
 #endif /* __KERNEL__ */
-- 
cgit v1.2.3-55-g7522


From ced2c8bcbc5082bc112d7c7410b3f5380c77e306 Mon Sep 17 00:00:00 2001
From: Hendrik Brueckner
Date: Thu, 26 Mar 2009 15:24:34 +0100
Subject: [S390] remove duplicate nul-termination of string

strlcpy() does already NUL-terminate the destination string.

Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/early.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c
index d61eb3334edf..4d221c81c849 100644
--- a/arch/s390/kernel/early.c
+++ b/arch/s390/kernel/early.c
@@ -352,7 +352,6 @@ static void __init setup_boot_command_line(void)
 
 	/* copy arch command line */
 	strlcpy(boot_command_line, COMMAND_LINE, ARCH_COMMAND_LINE_SIZE);
-	boot_command_line[ARCH_COMMAND_LINE_SIZE - 1] = 0;
 
 	/* append IPL PARM data to the boot command line */
 	if (MACHINE_IS_VM) {
-- 
cgit v1.2.3-55-g7522


From e13ed9b2704487e98d9241282765869fbd9a2dda Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:35 +0100
Subject: [S390] bitops: remove likely annotations

likely/unlikely profiling revealed that none of the branches in bitops
is taken likely or unlikely. So remove the annotations.
In addition the generated code is shorter.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/bitops.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/bitops.h b/arch/s390/include/asm/bitops.h
index 334964a0fb14..b30606f6d523 100644
--- a/arch/s390/include/asm/bitops.h
+++ b/arch/s390/include/asm/bitops.h
@@ -525,16 +525,16 @@ static inline unsigned long __ffs_word_loop(const unsigned long *addr,
 static inline unsigned long __ffz_word(unsigned long nr, unsigned long word)
 {
 #ifdef __s390x__
-	if (likely((word & 0xffffffff) == 0xffffffff)) {
+	if ((word & 0xffffffff) == 0xffffffff) {
 		word >>= 32;
 		nr += 32;
 	}
 #endif
-	if (likely((word & 0xffff) == 0xffff)) {
+	if ((word & 0xffff) == 0xffff) {
 		word >>= 16;
 		nr += 16;
 	}
-	if (likely((word & 0xff) == 0xff)) {
+	if ((word & 0xff) == 0xff) {
 		word >>= 8;
 		nr += 8;
 	}
@@ -549,16 +549,16 @@ static inline unsigned long __ffz_word(unsigned long nr, unsigned long word)
 static inline unsigned long __ffs_word(unsigned long nr, unsigned long word)
 {
 #ifdef __s390x__
-	if (likely((word & 0xffffffff) == 0)) {
+	if ((word & 0xffffffff) == 0) {
 		word >>= 32;
 		nr += 32;
 	}
 #endif
-	if (likely((word & 0xffff) == 0)) {
+	if ((word & 0xffff) == 0) {
 		word >>= 16;
 		nr += 16;
 	}
-	if (likely((word & 0xff) == 0)) {
+	if ((word & 0xff) == 0) {
 		word >>= 8;
 		nr += 8;
 	}
-- 
cgit v1.2.3-55-g7522


From 504665a91498f43d220b7d0942281067283a35f7 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky
Date: Thu, 26 Mar 2009 15:24:36 +0100
Subject: [S390] module function call optimization

Avoid the detour over the PLT if the branch target of a function call
in a module is in the range of the bras (16-bit) or brasl (32-bit)
instruction. The PLT is still generated but it is unused.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/module.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c
index 59b4e796680a..eed4a00cb676 100644
--- a/arch/s390/kernel/module.c
+++ b/arch/s390/kernel/module.c
@@ -310,15 +310,20 @@ apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
 			info->plt_initialized = 1;
 		}
 		if (r_type == R_390_PLTOFF16 ||
-		    r_type == R_390_PLTOFF32
-		    || r_type == R_390_PLTOFF64
-			)
+		    r_type == R_390_PLTOFF32 ||
+		    r_type == R_390_PLTOFF64)
 			val = me->arch.plt_offset - me->arch.got_offset +
 				info->plt_offset + rela->r_addend;
-		else
-			val =  (Elf_Addr) me->module_core +
-				me->arch.plt_offset + info->plt_offset + 
-				rela->r_addend - loc;
+		else {
+			if (!((r_type == R_390_PLT16DBL &&
+			       val - loc + 0xffffUL < 0x1ffffeUL) ||
+			      (r_type == R_390_PLT32DBL &&
+			       val - loc + 0xffffffffULL < 0x1fffffffeULL)))
+				val = (Elf_Addr) me->module_core +
+					me->arch.plt_offset +
+					info->plt_offset;
+			val += rela->r_addend - loc;
+		}
 		if (r_type == R_390_PLT16DBL)
 			*(unsigned short *) loc = val >> 1;
 		else if (r_type == R_390_PLTOFF16)
-- 
cgit v1.2.3-55-g7522


From 1edad85b16fdda43c8ab809e2779e8bf64ab8bb2 Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:37 +0100
Subject: [S390] use compiler builtin versions of strlen/strcpy/strcat

Use builtin variants if gcc 4 or newer is used to compile the kernel.
Generates better code than the asm variants.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/string.h | 8 ++++++++
 arch/s390/lib/string.c         | 8 ++++++++
 2 files changed, 16 insertions(+)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/string.h b/arch/s390/include/asm/string.h
index adf079170aa6..cd0241db5a46 100644
--- a/arch/s390/include/asm/string.h
+++ b/arch/s390/include/asm/string.h
@@ -100,6 +100,7 @@ static inline char *strcat(char *dst, const char *src)
 
 static inline char *strcpy(char *dst, const char *src)
 {
+#if __GNUC__ < 4
 	register int r0 asm("0") = 0;
 	char *ret = dst;
 
@@ -109,10 +110,14 @@ static inline char *strcpy(char *dst, const char *src)
 		: "+&a" (dst), "+&a" (src) : "d" (r0)
 		: "cc", "memory");
 	return ret;
+#else
+	return __builtin_strcpy(dst, src);
+#endif
 }
 
 static inline size_t strlen(const char *s)
 {
+#if __GNUC__ < 4
 	register unsigned long r0 asm("0") = 0;
 	const char *tmp = s;
 
@@ -121,6 +126,9 @@ static inline size_t strlen(const char *s)
 		"	jo	0b"
 		: "+d" (r0), "+a" (tmp) :  : "cc");
 	return r0 - (unsigned long) s;
+#else
+	return __builtin_strlen(s);
+#endif
 }
 
 static inline size_t strnlen(const char * s, size_t n)
diff --git a/arch/s390/lib/string.c b/arch/s390/lib/string.c
index ae5cf5d03d41..4143b7c19096 100644
--- a/arch/s390/lib/string.c
+++ b/arch/s390/lib/string.c
@@ -44,7 +44,11 @@ static inline char *__strnend(const char *s, size_t n)
  */
 size_t strlen(const char *s)
 {
+#if __GNUC__ < 4
 	return __strend(s) - s;
+#else
+	return __builtin_strlen(s);
+#endif
 }
 EXPORT_SYMBOL(strlen);
 
@@ -70,6 +74,7 @@ EXPORT_SYMBOL(strnlen);
  */
 char *strcpy(char *dest, const char *src)
 {
+#if __GNUC__ < 4
 	register int r0 asm("0") = 0;
 	char *ret = dest;
 
@@ -78,6 +83,9 @@ char *strcpy(char *dest, const char *src)
 		      : "+&a" (dest), "+&a" (src) : "d" (r0)
 		      : "cc", "memory" );
 	return ret;
+#else
+	return __builtin_strcpy(dest, src);
+#endif
 }
 EXPORT_SYMBOL(strcpy);
 
-- 
cgit v1.2.3-55-g7522


From eaf1b6fbca8d9be87bef2eafaa3f40bffe26ce04 Mon Sep 17 00:00:00 2001
From: Dan Smith
Date: Thu, 26 Mar 2009 15:24:40 +0100
Subject: [S390] Expose a constant for the number of words representing the CRs

We need to use this value in the checkpoint/restart code and would like to
have a constant instead of a magic '3'.

Cc: linux-s390@vger.kernel.org
Signed-off-by: Dan Smith <danms@us.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/ptrace.h   | 4 +++-
 arch/s390/kernel/compat_ptrace.h | 3 ++-
 2 files changed, 5 insertions(+), 2 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/ptrace.h b/arch/s390/include/asm/ptrace.h
index 8920025c3c02..f1b051630c50 100644
--- a/arch/s390/include/asm/ptrace.h
+++ b/arch/s390/include/asm/ptrace.h
@@ -172,6 +172,8 @@
 #define NUM_CRS		16
 #define NUM_ACRS	16
 
+#define NUM_CR_WORDS	3
+
 #define FPR_SIZE	8
 #define FPC_SIZE	4
 #define FPC_PAD_SIZE	4 /* gcc insists on aligning the fpregs */
@@ -334,7 +336,7 @@ struct pt_regs
  */
 typedef struct
 {
-	unsigned long cr[3];
+	unsigned long cr[NUM_CR_WORDS];
 } per_cr_words;
 
 #define PER_EM_MASK 0xE8000000UL
diff --git a/arch/s390/kernel/compat_ptrace.h b/arch/s390/kernel/compat_ptrace.h
index a2be3a978d5c..123dd660d7fb 100644
--- a/arch/s390/kernel/compat_ptrace.h
+++ b/arch/s390/kernel/compat_ptrace.h
@@ -1,10 +1,11 @@
 #ifndef _PTRACE32_H
 #define _PTRACE32_H
 
+#include <asm/ptrace.h>    /* needed for NUM_CR_WORDS */
 #include "compat_linux.h"  /* needed for psw_compat_t */
 
 typedef struct {
-	__u32 cr[3];
+	__u32 cr[NUM_CR_WORDS];
 } per_cr_words32;
 
 typedef struct {
-- 
cgit v1.2.3-55-g7522


From 2938af534d47891ddbced552e5d29f7b90bec609 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger
Date: Thu, 26 Mar 2009 15:24:41 +0100
Subject: [S390] Fix comments in lowcore structure

This patch fixes two addresses in the comments for the
lowcore structure. Looks like an copy-paste bug.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/lowcore.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h
index ee4b10ff9387..f94386ece0d3 100644
--- a/arch/s390/include/asm/lowcore.h
+++ b/arch/s390/include/asm/lowcore.h
@@ -315,8 +315,8 @@ struct _lowcore
 	__u8         op_access_id;             /* 0x0a2 */
 	__u8         ar_access_id;             /* 0x0a3 */
 	__u8         pad2[0xA8-0xA4];          /* 0x0a4 */
-	addr_t       trans_exc_code;           /* 0x0A0 */
-	addr_t       monitor_code;             /* 0x09c */
+	addr_t	     trans_exc_code;	       /* 0x0a8 */
+	addr_t	     monitor_code;	       /* 0x0b0 */
 	__u16        subchannel_id;            /* 0x0b8 */
 	__u16        subchannel_nr;            /* 0x0ba */
 	__u32        io_int_parm;              /* 0x0bc */
-- 
cgit v1.2.3-55-g7522


From 7b4684880dfc6c45bc56039ca5eada771d7643ab Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky
Date: Thu, 26 Mar 2009 15:24:42 +0100
Subject: [S390] eliminate cpuinfo_S390 structure

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/lowcore.h   | 15 +++++----
 arch/s390/include/asm/processor.h | 16 +--------
 arch/s390/include/asm/smp.h       |  7 +---
 arch/s390/kernel/processor.c      | 69 +++++++++++++++++++--------------------
 arch/s390/kernel/setup.c          | 10 ++----
 arch/s390/kernel/smp.c            |  8 ++---
 drivers/s390/cio/css.c            |  2 +-
 7 files changed, 52 insertions(+), 75 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h
index f94386ece0d3..ad543c11826d 100644
--- a/arch/s390/include/asm/lowcore.h
+++ b/arch/s390/include/asm/lowcore.h
@@ -80,7 +80,6 @@
 #define __LC_USER_ASCE			0xC50
 #define __LC_PANIC_STACK                0xC54
 #define __LC_CPUID                      0xC60
-#define __LC_CPUADDR                    0xC68
 #define __LC_IPLDEV                     0xC7C
 #define __LC_CURRENT			0xC90
 #define __LC_INT_CLOCK			0xC98
@@ -102,7 +101,6 @@
 #define __LC_USER_ASCE			0xD60
 #define __LC_PANIC_STACK                0xD68
 #define __LC_CPUID			0xD80
-#define __LC_CPUADDR			0xD88
 #define __LC_IPLDEV                     0xDB8
 #define __LC_CURRENT			0xDD8
 #define __LC_INT_CLOCK			0xDE8
@@ -273,8 +271,10 @@ struct _lowcore
 	__u32	     user_exec_asce;	       /* 0xc58 */
 	__u8	     pad10[0xc60-0xc5c];       /* 0xc5c */
 	/* entry.S sensitive area start */
-	struct       cpuinfo_S390 cpu_data;    /* 0xc60 */
-	__u32        ipl_device;               /* 0xc7c */
+	cpuid_t      cpu_id;		       /* 0xc60 */
+	__u32	     cpu_nr;		       /* 0xc68 */
+	__u32	     ipl_device;	       /* 0xc6c */
+	__u8	     pad_0xc70[0xc80-0xc70];   /* 0xc70 */
 	/* entry.S sensitive area end */
 
         /* SMP info area: defined by DJB */
@@ -366,9 +366,10 @@ struct _lowcore
 	__u64	     user_exec_asce;	       /* 0xd70 */
 	__u8	     pad10[0xd80-0xd78];       /* 0xd78 */
 	/* entry.S sensitive area start */
-	struct       cpuinfo_S390 cpu_data;    /* 0xd80 */
-	__u32        ipl_device;               /* 0xdb8 */
-	__u32        pad11;                    /* 0xdbc */
+	cpuid_t      cpu_id;		       /* 0xd80 */
+	__u32	     cpu_nr;		       /* 0xd88 */
+	__u32	     ipl_device;	       /* 0xd8c */
+	__u8	     pad_0xd90[0xdc0-0xd90];   /* 0xd90 */
 	/* entry.S sensitive area end */
 
         /* SMP info area: defined by DJB */
diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h
index db4523fe38ac..61862b3ac794 100644
--- a/arch/s390/include/asm/processor.h
+++ b/arch/s390/include/asm/processor.h
@@ -42,22 +42,8 @@ static inline void get_cpu_id(cpuid_t *ptr)
 	asm volatile("stidp 0(%1)" : "=m" (*ptr) : "a" (ptr));
 }
 
-struct cpuinfo_S390
-{
-        cpuid_t  cpu_id;
-        __u16    cpu_addr;
-        __u16    cpu_nr;
-        unsigned long loops_per_jiffy;
-        unsigned long *pgd_quick;
-#ifdef __s390x__
-        unsigned long *pmd_quick;
-#endif /* __s390x__ */
-        unsigned long *pte_quick;
-        unsigned long pgtable_cache_sz;
-};
-
 extern void s390_adjust_jiffies(void);
-extern void print_cpu_info(struct cpuinfo_S390 *);
+extern void print_cpu_info(void);
 extern int get_cpu_capability(unsigned int *);
 
 /*
diff --git a/arch/s390/include/asm/smp.h b/arch/s390/include/asm/smp.h
index 024b91e06239..2009158a4502 100644
--- a/arch/s390/include/asm/smp.h
+++ b/arch/s390/include/asm/smp.h
@@ -50,12 +50,7 @@ extern void machine_power_off_smp(void);
  
 #define PROC_CHANGE_PENALTY	20		/* Schedule penalty */
 
-#define raw_smp_processor_id()	(S390_lowcore.cpu_data.cpu_nr)
-
-static inline __u16 hard_smp_processor_id(void)
-{
-	return stap();
-}
+#define raw_smp_processor_id()	(S390_lowcore.cpu_nr)
 
 /*
  * returns 1 if cpu is in stopped/check stopped state or not operational
diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c
index 82c1872cfe80..423da1bd42a4 100644
--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -18,10 +18,11 @@
 #include <asm/lowcore.h>
 #include <asm/param.h>
 
-void __cpuinit print_cpu_info(struct cpuinfo_S390 *cpuinfo)
+void __cpuinit print_cpu_info(void)
 {
 	pr_info("Processor %d started, address %d, identification %06X\n",
-		cpuinfo->cpu_nr, cpuinfo->cpu_addr, cpuinfo->cpu_id.ident);
+		S390_lowcore.cpu_nr, S390_lowcore.cpu_addr,
+		S390_lowcore.cpu_id.ident);
 }
 
 /*
@@ -34,44 +35,42 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		"esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp",
 		"edat"
 	};
-       struct cpuinfo_S390 *cpuinfo;
-       unsigned long n = (unsigned long) v - 1;
-       int i;
+	struct _lowcore *lc;
+	unsigned long n = (unsigned long) v - 1;
+	int i;
 
-       s390_adjust_jiffies();
-       preempt_disable();
-       if (!n) {
-	       seq_printf(m, "vendor_id       : IBM/S390\n"
-			  "# processors    : %i\n"
-			  "bogomips per cpu: %lu.%02lu\n",
-			  num_online_cpus(), loops_per_jiffy/(500000/HZ),
-			  (loops_per_jiffy/(5000/HZ))%100);
-	       seq_puts(m, "features\t: ");
-	       for (i = 0; i < 8; i++)
-		       if (hwcap_str[i] && (elf_hwcap & (1UL << i)))
-			       seq_printf(m, "%s ", hwcap_str[i]);
-	       seq_puts(m, "\n");
-       }
+	s390_adjust_jiffies();
+	preempt_disable();
+	if (!n) {
+		seq_printf(m, "vendor_id       : IBM/S390\n"
+			   "# processors    : %i\n"
+			   "bogomips per cpu: %lu.%02lu\n",
+			   num_online_cpus(), loops_per_jiffy/(500000/HZ),
+			   (loops_per_jiffy/(5000/HZ))%100);
+		seq_puts(m, "features\t: ");
+		for (i = 0; i < 8; i++)
+			if (hwcap_str[i] && (elf_hwcap & (1UL << i)))
+				seq_printf(m, "%s ", hwcap_str[i]);
+		seq_puts(m, "\n");
+	}
 
-       if (cpu_online(n)) {
+	if (cpu_online(n)) {
 #ifdef CONFIG_SMP
-	       if (smp_processor_id() == n)
-		       cpuinfo = &S390_lowcore.cpu_data;
-	       else
-		       cpuinfo = &lowcore_ptr[n]->cpu_data;
+		lc = (smp_processor_id() == n) ?
+			&S390_lowcore : lowcore_ptr[n];
 #else
-	       cpuinfo = &S390_lowcore.cpu_data;
+		lc = &S390_lowcore;
 #endif
-	       seq_printf(m, "processor %li: "
-			  "version = %02X,  "
-			  "identification = %06X,  "
-			  "machine = %04X\n",
-			  n, cpuinfo->cpu_id.version,
-			  cpuinfo->cpu_id.ident,
-			  cpuinfo->cpu_id.machine);
-       }
-       preempt_enable();
-       return 0;
+		seq_printf(m, "processor %li: "
+			   "version = %02X,  "
+			   "identification = %06X,  "
+			   "machine = %04X\n",
+			   n, lc->cpu_id.version,
+			   lc->cpu_id.ident,
+			   lc->cpu_id.machine);
+	}
+	preempt_enable();
+	return 0;
 }
 
 static void *c_start(struct seq_file *m, loff_t *pos)
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index dd3c51736270..9c8853f21bb2 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -121,13 +121,10 @@ static struct resource data_resource = {
  */
 void __cpuinit cpu_init(void)
 {
-        int addr = hard_smp_processor_id();
-
         /*
          * Store processor id in lowcore (used e.g. in timer_interrupt)
          */
-	get_cpu_id(&S390_lowcore.cpu_data.cpu_id);
-        S390_lowcore.cpu_data.cpu_addr = addr;
+	get_cpu_id(&S390_lowcore.cpu_id);
 
         /*
          * Force FPU initialization:
@@ -686,7 +683,6 @@ setup_memory(void)
 static void __init setup_hwcaps(void)
 {
 	static const int stfl_bits[6] = { 0, 2, 7, 17, 19, 21 };
-	struct cpuinfo_S390 *cpuinfo = &S390_lowcore.cpu_data;
 	unsigned long long facility_list_extended;
 	unsigned int facility_list;
 	int i;
@@ -732,7 +728,7 @@ static void __init setup_hwcaps(void)
 	if (MACHINE_HAS_HPAGE)
 		elf_hwcap |= 1UL << 7;
 
-	switch (cpuinfo->cpu_id.machine) {
+	switch (S390_lowcore.cpu_id.machine) {
 	case 0x9672:
 #if !defined(CONFIG_64BIT)
 	default:	/* Use "g5" as default for 31 bit kernels. */
@@ -825,7 +821,7 @@ setup_arch(char **cmdline_p)
 	setup_lowcore();
 
         cpu_init();
-        __cpu_logical_map[0] = S390_lowcore.cpu_data.cpu_addr;
+	__cpu_logical_map[0] = stap();
 	s390_init_cpu_topology();
 
 	/*
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index a8858634dd05..b167f74d94cd 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -372,7 +372,7 @@ static void __init smp_detect_cpus(void)
 
 	c_cpus = 1;
 	s_cpus = 0;
-	boot_cpu_addr = S390_lowcore.cpu_data.cpu_addr;
+	boot_cpu_addr = __cpu_logical_map[0];
 	info = kmalloc(sizeof(*info), GFP_KERNEL);
 	if (!info)
 		panic("smp_detect_cpus failed to allocate memory\n");
@@ -446,7 +446,7 @@ int __cpuinit start_secondary(void *cpuvoid)
 	/* Switch on interrupts */
 	local_irq_enable();
 	/* Print info about this processor */
-	print_cpu_info(&S390_lowcore.cpu_data);
+	print_cpu_info();
 	/* cpu_idle will call schedule for us */
 	cpu_idle();
 	return 0;
@@ -564,7 +564,7 @@ int __cpuinit __cpu_up(unsigned int cpu)
 		: : "a" (&cpu_lowcore->access_regs_save_area) : "memory");
 	cpu_lowcore->percpu_offset = __per_cpu_offset[cpu];
 	cpu_lowcore->current_task = (unsigned long) idle;
-	cpu_lowcore->cpu_data.cpu_nr = cpu;
+	cpu_lowcore->cpu_nr = cpu;
 	cpu_lowcore->kernel_asce = S390_lowcore.kernel_asce;
 	cpu_lowcore->ipl_device = S390_lowcore.ipl_device;
 	eieio();
@@ -656,7 +656,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	/* request the 0x1201 emergency signal external interrupt */
 	if (register_external_interrupt(0x1201, do_ext_call_interrupt) != 0)
 		panic("Couldn't request external interrupt 0x1201");
-	print_cpu_info(&S390_lowcore.cpu_data);
+	print_cpu_info();
 
 	/* Reallocate current lowcore, but keep its contents. */
 	lc_order = sizeof(long) == 8 ? 1 : 0;
diff --git a/drivers/s390/cio/css.c b/drivers/s390/cio/css.c
index 8446d15e4485..dcd0e48918da 100644
--- a/drivers/s390/cio/css.c
+++ b/drivers/s390/cio/css.c
@@ -655,7 +655,7 @@ css_generate_pgid(struct channel_subsystem *css, u32 tod_high)
 		css->global_pgid.pgid_high.ext_cssid.cssid = css->cssid;
 	} else {
 #ifdef CONFIG_SMP
-		css->global_pgid.pgid_high.cpu_addr = hard_smp_processor_id();
+		css->global_pgid.pgid_high.cpu_addr = stap();
 #else
 		css->global_pgid.pgid_high.cpu_addr = 0;
 #endif
-- 
cgit v1.2.3-55-g7522


From da292bbe1f620221b08c4b589424f370168d642b Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky
Date: Thu, 26 Mar 2009 15:24:43 +0100
Subject: [S390] eliminate ipl_device from lowcore

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/lowcore.h | 8 ++------
 arch/s390/kernel/head31.S       | 1 -
 arch/s390/kernel/head64.S       | 1 -
 arch/s390/kernel/setup.c        | 1 -
 arch/s390/kernel/smp.c          | 1 -
 5 files changed, 2 insertions(+), 10 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h
index ad543c11826d..5b18035c1dc7 100644
--- a/arch/s390/include/asm/lowcore.h
+++ b/arch/s390/include/asm/lowcore.h
@@ -80,7 +80,6 @@
 #define __LC_USER_ASCE			0xC50
 #define __LC_PANIC_STACK                0xC54
 #define __LC_CPUID                      0xC60
-#define __LC_IPLDEV                     0xC7C
 #define __LC_CURRENT			0xC90
 #define __LC_INT_CLOCK			0xC98
 #else /* __s390x__ */
@@ -101,7 +100,6 @@
 #define __LC_USER_ASCE			0xD60
 #define __LC_PANIC_STACK                0xD68
 #define __LC_CPUID			0xD80
-#define __LC_IPLDEV                     0xDB8
 #define __LC_CURRENT			0xDD8
 #define __LC_INT_CLOCK			0xDE8
 #define __LC_VDSO_PER_CPU		0xE38
@@ -273,8 +271,7 @@ struct _lowcore
 	/* entry.S sensitive area start */
 	cpuid_t      cpu_id;		       /* 0xc60 */
 	__u32	     cpu_nr;		       /* 0xc68 */
-	__u32	     ipl_device;	       /* 0xc6c */
-	__u8	     pad_0xc70[0xc80-0xc70];   /* 0xc70 */
+	__u8	     pad_0xc6c[0xc80-0xc6c];   /* 0xc6c */
 	/* entry.S sensitive area end */
 
         /* SMP info area: defined by DJB */
@@ -368,8 +365,7 @@ struct _lowcore
 	/* entry.S sensitive area start */
 	cpuid_t      cpu_id;		       /* 0xd80 */
 	__u32	     cpu_nr;		       /* 0xd88 */
-	__u32	     ipl_device;	       /* 0xd8c */
-	__u8	     pad_0xd90[0xdc0-0xd90];   /* 0xd90 */
+	__u8	     pad_0xd8c[0xdc0-0xd8c];   /* 0xd8c */
 	/* entry.S sensitive area end */
 
         /* SMP info area: defined by DJB */
diff --git a/arch/s390/kernel/head31.S b/arch/s390/kernel/head31.S
index db476d114caa..2ced846065b7 100644
--- a/arch/s390/kernel/head31.S
+++ b/arch/s390/kernel/head31.S
@@ -20,7 +20,6 @@ startup_continue:
 	lctl	%c0,%c15,.Lctl-.LPG1(%r13) # load control registers
 	l	%r12,.Lparmaddr-.LPG1(%r13) # pointer to parameter area
 					# move IPL device to lowcore
-	mvc	__LC_IPLDEV(4),IPL_DEVICE-PARMAREA(%r12)
 #
 # Setup stack
 #
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index f9f70aa15244..65667b2e65ce 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -86,7 +86,6 @@ startup_continue:
 	lctlg	%c0,%c15,.Lctl-.LPG1(%r13)	# load control registers
 	lg	%r12,.Lparmaddr-.LPG1(%r13)	# pointer to parameter area
 					# move IPL device to lowcore
-	mvc	__LC_IPLDEV(4),IPL_DEVICE+4-PARMAREA(%r12)
 	lghi	%r0,__LC_PASTE
 	stg	%r0,__LC_VDSO_PER_CPU
 #
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 9c8853f21bb2..91551ef1d67e 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -419,7 +419,6 @@ setup_lowcore(void)
 		PSW_ADDR_AMODE | (unsigned long) mcck_int_handler;
 	lc->io_new_psw.mask = psw_kernel_bits;
 	lc->io_new_psw.addr = PSW_ADDR_AMODE | (unsigned long) io_int_handler;
-	lc->ipl_device = S390_lowcore.ipl_device;
 	lc->clock_comparator = -1ULL;
 	lc->kernel_stack = ((unsigned long) &init_thread_union) + THREAD_SIZE;
 	lc->async_stack = (unsigned long)
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index b167f74d94cd..775885772ff1 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -566,7 +566,6 @@ int __cpuinit __cpu_up(unsigned int cpu)
 	cpu_lowcore->current_task = (unsigned long) idle;
 	cpu_lowcore->cpu_nr = cpu;
 	cpu_lowcore->kernel_asce = S390_lowcore.kernel_asce;
-	cpu_lowcore->ipl_device = S390_lowcore.ipl_device;
 	eieio();
 
 	while (signal_processor(cpu, sigp_restart) == sigp_busy)
-- 
cgit v1.2.3-55-g7522


From 866ba28418d30122d863c50182a202741f4dcf3e Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky
Date: Thu, 26 Mar 2009 15:24:44 +0100
Subject: [S390] cleanup lowcore.h

The lowcore.h header has quite a lot of whitespace damage and a rather
wild collection of entries. Remove all that whitespace and tidy up the
order of the lowcore fields.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/lowcore.h | 652 ++++++++++++++++++++--------------------
 arch/s390/kernel/head.S         |   2 +
 2 files changed, 331 insertions(+), 323 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/lowcore.h b/arch/s390/include/asm/lowcore.h
index 5b18035c1dc7..b349f1c7fdfa 100644
--- a/arch/s390/include/asm/lowcore.h
+++ b/arch/s390/include/asm/lowcore.h
@@ -11,125 +11,118 @@
 #ifndef _ASM_S390_LOWCORE_H
 #define _ASM_S390_LOWCORE_H
 
-#ifndef __s390x__
-#define __LC_EXT_OLD_PSW                0x018
-#define __LC_SVC_OLD_PSW                0x020
-#define __LC_PGM_OLD_PSW                0x028
-#define __LC_MCK_OLD_PSW                0x030
-#define __LC_IO_OLD_PSW                 0x038
-#define __LC_EXT_NEW_PSW                0x058
-#define __LC_SVC_NEW_PSW                0x060
-#define __LC_PGM_NEW_PSW                0x068
-#define __LC_MCK_NEW_PSW                0x070
-#define __LC_IO_NEW_PSW                 0x078
-#else /* !__s390x__ */
-#define __LC_EXT_OLD_PSW                0x0130
-#define __LC_SVC_OLD_PSW                0x0140
-#define __LC_PGM_OLD_PSW                0x0150
-#define __LC_MCK_OLD_PSW                0x0160
-#define __LC_IO_OLD_PSW                 0x0170
-#define __LC_EXT_NEW_PSW                0x01b0
-#define __LC_SVC_NEW_PSW                0x01c0
-#define __LC_PGM_NEW_PSW                0x01d0
-#define __LC_MCK_NEW_PSW                0x01e0
-#define __LC_IO_NEW_PSW                 0x01f0
-#endif /* !__s390x__ */
-
-#define __LC_IPL_PARMBLOCK_PTR		0x014
-#define __LC_EXT_PARAMS                 0x080
-#define __LC_CPU_ADDRESS                0x084
-#define __LC_EXT_INT_CODE               0x086
-
-#define __LC_SVC_ILC                    0x088
-#define __LC_SVC_INT_CODE               0x08A
-#define __LC_PGM_ILC                    0x08C
-#define __LC_PGM_INT_CODE               0x08E
+#define __LC_IPL_PARMBLOCK_PTR		0x0014
+#define __LC_EXT_PARAMS			0x0080
+#define __LC_CPU_ADDRESS		0x0084
+#define __LC_EXT_INT_CODE		0x0086
 
-#define __LC_PER_ATMID			0x096
-#define __LC_PER_ADDRESS		0x098
-#define __LC_PER_ACCESS_ID		0x0A1
-#define __LC_AR_MODE_ID			0x0A3
+#define __LC_SVC_ILC			0x0088
+#define __LC_SVC_INT_CODE		0x008a
+#define __LC_PGM_ILC			0x008c
+#define __LC_PGM_INT_CODE		0x008e
 
-#define __LC_SUBCHANNEL_ID              0x0B8
-#define __LC_SUBCHANNEL_NR              0x0BA
-#define __LC_IO_INT_PARM                0x0BC
-#define __LC_IO_INT_WORD                0x0C0
-#define __LC_MCCK_CODE                  0x0E8
+#define __LC_PER_ATMID			0x0096
+#define __LC_PER_ADDRESS		0x0098
+#define __LC_PER_ACCESS_ID		0x00a1
+#define __LC_AR_MODE_ID			0x00a3
 
-#define __LC_LAST_BREAK 		0x110
+#define __LC_SUBCHANNEL_ID		0x00b8
+#define __LC_SUBCHANNEL_NR		0x00ba
+#define __LC_IO_INT_PARM		0x00bc
+#define __LC_IO_INT_WORD		0x00c0
+#define __LC_MCCK_CODE			0x00e8
 
-#define __LC_RETURN_PSW                 0x200
-
-#define __LC_SAVE_AREA                  0xC00
+#define __LC_DUMP_REIPL			0x0e00
 
 #ifndef __s390x__
-#define __LC_IRB			0x208
-#define __LC_SYNC_ENTER_TIMER		0x248
-#define __LC_ASYNC_ENTER_TIMER		0x250
-#define __LC_EXIT_TIMER			0x258
-#define __LC_USER_TIMER			0x260
-#define __LC_SYSTEM_TIMER		0x268
-#define __LC_STEAL_TIMER		0x270
-#define __LC_LAST_UPDATE_TIMER		0x278
-#define __LC_LAST_UPDATE_CLOCK		0x280
-#define __LC_RETURN_MCCK_PSW            0x288
-#define __LC_KERNEL_STACK               0xC40
-#define __LC_THREAD_INFO		0xC44
-#define __LC_ASYNC_STACK                0xC48
-#define __LC_KERNEL_ASCE		0xC4C
-#define __LC_USER_ASCE			0xC50
-#define __LC_PANIC_STACK                0xC54
-#define __LC_CPUID                      0xC60
-#define __LC_CURRENT			0xC90
-#define __LC_INT_CLOCK			0xC98
+#define __LC_EXT_OLD_PSW		0x0018
+#define __LC_SVC_OLD_PSW		0x0020
+#define __LC_PGM_OLD_PSW		0x0028
+#define __LC_MCK_OLD_PSW		0x0030
+#define __LC_IO_OLD_PSW			0x0038
+#define __LC_EXT_NEW_PSW		0x0058
+#define __LC_SVC_NEW_PSW		0x0060
+#define __LC_PGM_NEW_PSW		0x0068
+#define __LC_MCK_NEW_PSW		0x0070
+#define __LC_IO_NEW_PSW			0x0078
+#define __LC_SAVE_AREA			0x0200
+#define __LC_RETURN_PSW			0x0240
+#define __LC_RETURN_MCCK_PSW		0x0248
+#define __LC_SYNC_ENTER_TIMER		0x0250
+#define __LC_ASYNC_ENTER_TIMER		0x0258
+#define __LC_EXIT_TIMER			0x0260
+#define __LC_USER_TIMER			0x0268
+#define __LC_SYSTEM_TIMER		0x0270
+#define __LC_STEAL_TIMER		0x0278
+#define __LC_LAST_UPDATE_TIMER		0x0280
+#define __LC_LAST_UPDATE_CLOCK		0x0288
+#define __LC_CURRENT			0x0290
+#define __LC_THREAD_INFO		0x0294
+#define __LC_KERNEL_STACK		0x0298
+#define __LC_ASYNC_STACK		0x029c
+#define __LC_PANIC_STACK		0x02a0
+#define __LC_KERNEL_ASCE		0x02a4
+#define __LC_USER_ASCE			0x02a8
+#define __LC_USER_EXEC_ASCE		0x02ac
+#define __LC_CPUID			0x02b0
+#define __LC_INT_CLOCK			0x02c8
+#define __LC_IRB			0x0300
+#define __LC_PFAULT_INTPARM		0x0080
+#define __LC_CPU_TIMER_SAVE_AREA	0x00d8
+#define __LC_CLOCK_COMP_SAVE_AREA	0x00e0
+#define __LC_PSW_SAVE_AREA		0x0100
+#define __LC_PREFIX_SAVE_AREA		0x0108
+#define __LC_AREGS_SAVE_AREA		0x0120
+#define __LC_FPREGS_SAVE_AREA		0x0160
+#define __LC_GPREGS_SAVE_AREA		0x0180
+#define __LC_CREGS_SAVE_AREA		0x01c0
 #else /* __s390x__ */
-#define __LC_IRB			0x210
-#define __LC_SYNC_ENTER_TIMER		0x250
-#define __LC_ASYNC_ENTER_TIMER		0x258
-#define __LC_EXIT_TIMER			0x260
-#define __LC_USER_TIMER			0x268
-#define __LC_SYSTEM_TIMER		0x270
-#define __LC_STEAL_TIMER		0x278
-#define __LC_LAST_UPDATE_TIMER		0x280
-#define __LC_LAST_UPDATE_CLOCK		0x288
-#define __LC_RETURN_MCCK_PSW            0x290
-#define __LC_KERNEL_STACK               0xD40
-#define __LC_THREAD_INFO		0xD48
-#define __LC_ASYNC_STACK                0xD50
-#define __LC_KERNEL_ASCE		0xD58
-#define __LC_USER_ASCE			0xD60
-#define __LC_PANIC_STACK                0xD68
-#define __LC_CPUID			0xD80
-#define __LC_CURRENT			0xDD8
-#define __LC_INT_CLOCK			0xDE8
-#define __LC_VDSO_PER_CPU		0xE38
-#endif /* __s390x__ */
-
-#define __LC_PASTE			0xE40
-
-#define __LC_DUMP_REIPL			0xE00
-#ifndef __s390x__
-#define __LC_PFAULT_INTPARM             0x080
-#define __LC_CPU_TIMER_SAVE_AREA        0x0D8
-#define __LC_CLOCK_COMP_SAVE_AREA	0x0E0
-#define __LC_PSW_SAVE_AREA		0x100
-#define __LC_PREFIX_SAVE_AREA		0x108
-#define __LC_AREGS_SAVE_AREA            0x120
-#define __LC_FPREGS_SAVE_AREA		0x160
-#define __LC_GPREGS_SAVE_AREA           0x180
-#define __LC_CREGS_SAVE_AREA            0x1C0
-#else /* __s390x__ */
-#define __LC_PFAULT_INTPARM             0x11B8
+#define __LC_LAST_BREAK			0x0110
+#define __LC_EXT_OLD_PSW		0x0130
+#define __LC_SVC_OLD_PSW		0x0140
+#define __LC_PGM_OLD_PSW		0x0150
+#define __LC_MCK_OLD_PSW		0x0160
+#define __LC_IO_OLD_PSW			0x0170
+#define __LC_EXT_NEW_PSW		0x01b0
+#define __LC_SVC_NEW_PSW		0x01c0
+#define __LC_PGM_NEW_PSW		0x01d0
+#define __LC_MCK_NEW_PSW		0x01e0
+#define __LC_IO_NEW_PSW			0x01f0
+#define __LC_SAVE_AREA			0x0200
+#define __LC_RETURN_PSW			0x0280
+#define __LC_RETURN_MCCK_PSW		0x0290
+#define __LC_SYNC_ENTER_TIMER		0x02a0
+#define __LC_ASYNC_ENTER_TIMER		0x02a8
+#define __LC_EXIT_TIMER			0x02b0
+#define __LC_USER_TIMER			0x02b8
+#define __LC_SYSTEM_TIMER		0x02c0
+#define __LC_STEAL_TIMER		0x02c8
+#define __LC_LAST_UPDATE_TIMER		0x02d0
+#define __LC_LAST_UPDATE_CLOCK		0x02d8
+#define __LC_CURRENT			0x02e0
+#define __LC_THREAD_INFO		0x02e8
+#define __LC_KERNEL_STACK		0x02f0
+#define __LC_ASYNC_STACK		0x02f8
+#define __LC_PANIC_STACK		0x0300
+#define __LC_KERNEL_ASCE		0x0308
+#define __LC_USER_ASCE			0x0310
+#define __LC_USER_EXEC_ASCE		0x0318
+#define __LC_CPUID			0x0320
+#define __LC_INT_CLOCK			0x0340
+#define __LC_VDSO_PER_CPU		0x0350
+#define __LC_IRB			0x0380
+#define __LC_PASTE			0x03c0
+#define __LC_PFAULT_INTPARM		0x11b8
 #define __LC_FPREGS_SAVE_AREA		0x1200
-#define __LC_GPREGS_SAVE_AREA           0x1280
+#define __LC_GPREGS_SAVE_AREA		0x1280
 #define __LC_PSW_SAVE_AREA		0x1300
 #define __LC_PREFIX_SAVE_AREA		0x1318
-#define __LC_FP_CREG_SAVE_AREA		0x131C
+#define __LC_FP_CREG_SAVE_AREA		0x131c
 #define __LC_TODREG_SAVE_AREA		0x1324
-#define __LC_CPU_TIMER_SAVE_AREA        0x1328
+#define __LC_CPU_TIMER_SAVE_AREA	0x1328
 #define __LC_CLOCK_COMP_SAVE_AREA	0x1331
-#define __LC_AREGS_SAVE_AREA            0x1340
-#define __LC_CREGS_SAVE_AREA            0x1380
+#define __LC_AREGS_SAVE_AREA		0x1340
+#define __LC_CREGS_SAVE_AREA		0x1380
 #endif /* __s390x__ */
 
 #ifndef __ASSEMBLY__
@@ -194,227 +187,240 @@ union save_area {
 struct _lowcore
 {
 #ifndef __s390x__
-        /* prefix area: defined by architecture */
-	psw_t        restart_psw;              /* 0x000 */
-	__u32        ccw2[4];                  /* 0x008 */
-	psw_t        external_old_psw;         /* 0x018 */
-	psw_t        svc_old_psw;              /* 0x020 */
-	psw_t        program_old_psw;          /* 0x028 */
-	psw_t        mcck_old_psw;             /* 0x030 */
-	psw_t        io_old_psw;               /* 0x038 */
-	__u8         pad1[0x58-0x40];          /* 0x040 */
-	psw_t        external_new_psw;         /* 0x058 */
-	psw_t        svc_new_psw;              /* 0x060 */
-	psw_t        program_new_psw;          /* 0x068 */
-	psw_t        mcck_new_psw;             /* 0x070 */
-	psw_t        io_new_psw;               /* 0x078 */
-	__u32        ext_params;               /* 0x080 */
-	__u16        cpu_addr;                 /* 0x084 */
-	__u16        ext_int_code;             /* 0x086 */
-        __u16        svc_ilc;                  /* 0x088 */
-        __u16        svc_code;                 /* 0x08a */
-        __u16        pgm_ilc;                  /* 0x08c */
-        __u16        pgm_code;                 /* 0x08e */
-	__u32        trans_exc_code;           /* 0x090 */
-	__u16        mon_class_num;            /* 0x094 */
-	__u16        per_perc_atmid;           /* 0x096 */
-	__u32        per_address;              /* 0x098 */
-	__u32        monitor_code;             /* 0x09c */
-	__u8         exc_access_id;            /* 0x0a0 */
-	__u8         per_access_id;            /* 0x0a1 */
-	__u8         pad2[0xB8-0xA2];          /* 0x0a2 */
-	__u16        subchannel_id;            /* 0x0b8 */
-	__u16        subchannel_nr;            /* 0x0ba */
-	__u32        io_int_parm;              /* 0x0bc */
-	__u32        io_int_word;              /* 0x0c0 */
-	__u8	     pad3[0xc8-0xc4];	       /* 0x0c4 */
-	__u32	     stfl_fac_list;	       /* 0x0c8 */
-	__u8	     pad4[0xd4-0xcc];	       /* 0x0cc */
-	__u32        extended_save_area_addr;  /* 0x0d4 */
-	__u32        cpu_timer_save_area[2];   /* 0x0d8 */
-	__u32        clock_comp_save_area[2];  /* 0x0e0 */
-	__u32        mcck_interruption_code[2]; /* 0x0e8 */
-	__u8	     pad5[0xf4-0xf0];	       /* 0x0f0 */
-	__u32        external_damage_code;     /* 0x0f4 */
-	__u32        failing_storage_address;  /* 0x0f8 */
-	__u8	     pad6[0x100-0xfc];	       /* 0x0fc */
-	__u32        st_status_fixed_logout[4];/* 0x100 */
-	__u8	     pad7[0x120-0x110];        /* 0x110 */
-	__u32        access_regs_save_area[16];/* 0x120 */
-	__u32        floating_pt_save_area[8]; /* 0x160 */
-	__u32        gpregs_save_area[16];     /* 0x180 */
-	__u32        cregs_save_area[16];      /* 0x1c0 */	
-
-        psw_t        return_psw;               /* 0x200 */
-	__u8	     irb[64];		       /* 0x208 */
-	__u64        sync_enter_timer;         /* 0x248 */
-	__u64        async_enter_timer;        /* 0x250 */
-	__u64        exit_timer;               /* 0x258 */
-	__u64	     user_timer;	       /* 0x260 */
-	__u64	     system_timer;	       /* 0x268 */
-	__u64	     steal_timer;	       /* 0x270 */
-	__u64	     last_update_timer;        /* 0x278 */
-	__u64	     last_update_clock;        /* 0x280 */
-        psw_t        return_mcck_psw;          /* 0x288 */
-	__u8         pad8[0xc00-0x290];        /* 0x290 */
-
-        /* System info area */
-	__u32        save_area[16];            /* 0xc00 */
-	__u32        kernel_stack;             /* 0xc40 */
-	__u32        thread_info;              /* 0xc44 */
-	__u32        async_stack;              /* 0xc48 */
-	__u32        kernel_asce;              /* 0xc4c */
-	__u32        user_asce;                /* 0xc50 */
-	__u32        panic_stack;              /* 0xc54 */
-	__u32	     user_exec_asce;	       /* 0xc58 */
-	__u8	     pad10[0xc60-0xc5c];       /* 0xc5c */
-	/* entry.S sensitive area start */
-	cpuid_t      cpu_id;		       /* 0xc60 */
-	__u32	     cpu_nr;		       /* 0xc68 */
-	__u8	     pad_0xc6c[0xc80-0xc6c];   /* 0xc6c */
-	/* entry.S sensitive area end */
-
-        /* SMP info area: defined by DJB */
-	__u64	     clock_comparator;	       /* 0xc80 */
-	__u32        ext_call_fast;            /* 0xc88 */
-	__u32        percpu_offset;            /* 0xc8c */
-	__u32        current_task;	       /* 0xc90 */
-	__u32        softirq_pending;	       /* 0xc94 */
-	__u64        int_clock;                /* 0xc98 */
-        __u8         pad11[0xe00-0xca0];       /* 0xca0 */
-
-	/* 0xe00 contains the address of the IPL Parameter */
-	/* Information block. Dump tools need IPIB for IPL */
-	/* after dump.					   */
-	__u32	     ipib;		       /* 0xe00 */
-	__u32	     ipib_checksum;	       /* 0xe04 */
-
-        /* Align to the top 1k of prefix area */
-	__u8	     pad12[0x1000-0xe08];      /* 0xe08 */
+	/* 0x0000 - 0x01ff: defined by architecture */
+	psw_t	restart_psw;			/* 0x0000 */
+	__u32	ccw2[4];			/* 0x0008 */
+	psw_t	external_old_psw;		/* 0x0018 */
+	psw_t	svc_old_psw;			/* 0x0020 */
+	psw_t	program_old_psw;		/* 0x0028 */
+	psw_t	mcck_old_psw;			/* 0x0030 */
+	psw_t	io_old_psw;			/* 0x0038 */
+	__u8	pad_0x0040[0x0058-0x0040];	/* 0x0040 */
+	psw_t	external_new_psw;		/* 0x0058 */
+	psw_t	svc_new_psw;			/* 0x0060 */
+	psw_t	program_new_psw;		/* 0x0068 */
+	psw_t	mcck_new_psw;			/* 0x0070 */
+	psw_t	io_new_psw;			/* 0x0078 */
+	__u32	ext_params;			/* 0x0080 */
+	__u16	cpu_addr;			/* 0x0084 */
+	__u16	ext_int_code;			/* 0x0086 */
+	__u16	svc_ilc;			/* 0x0088 */
+	__u16	svc_code;			/* 0x008a */
+	__u16	pgm_ilc;			/* 0x008c */
+	__u16	pgm_code;			/* 0x008e */
+	__u32	trans_exc_code;			/* 0x0090 */
+	__u16	mon_class_num;			/* 0x0094 */
+	__u16	per_perc_atmid;			/* 0x0096 */
+	__u32	per_address;			/* 0x0098 */
+	__u32	monitor_code;			/* 0x009c */
+	__u8	exc_access_id;			/* 0x00a0 */
+	__u8	per_access_id;			/* 0x00a1 */
+	__u8	pad_0x00a2[0x00b8-0x00a2];	/* 0x00a2 */
+	__u16	subchannel_id;			/* 0x00b8 */
+	__u16	subchannel_nr;			/* 0x00ba */
+	__u32	io_int_parm;			/* 0x00bc */
+	__u32	io_int_word;			/* 0x00c0 */
+	__u8	pad_0x00c4[0x00c8-0x00c4];	/* 0x00c4 */
+	__u32	stfl_fac_list;			/* 0x00c8 */
+	__u8	pad_0x00cc[0x00d4-0x00cc];	/* 0x00cc */
+	__u32	extended_save_area_addr;	/* 0x00d4 */
+	__u32	cpu_timer_save_area[2];		/* 0x00d8 */
+	__u32	clock_comp_save_area[2];	/* 0x00e0 */
+	__u32	mcck_interruption_code[2];	/* 0x00e8 */
+	__u8	pad_0x00f0[0x00f4-0x00f0];	/* 0x00f0 */
+	__u32	external_damage_code;		/* 0x00f4 */
+	__u32	failing_storage_address;	/* 0x00f8 */
+	__u8	pad_0x00fc[0x0100-0x00fc];	/* 0x00fc */
+	__u32	st_status_fixed_logout[4];	/* 0x0100 */
+	__u8	pad_0x0110[0x0120-0x0110];	/* 0x0110 */
+
+	/* CPU register save area: defined by architecture */
+	__u32	access_regs_save_area[16];	/* 0x0120 */
+	__u32	floating_pt_save_area[8];	/* 0x0160 */
+	__u32	gpregs_save_area[16];		/* 0x0180 */
+	__u32	cregs_save_area[16];		/* 0x01c0 */
+
+	/* Return psws. */
+	__u32	save_area[16];			/* 0x0200 */
+	psw_t	return_psw;			/* 0x0240 */
+	psw_t	return_mcck_psw;		/* 0x0248 */
+
+	/* CPU time accounting values */
+	__u64	sync_enter_timer;		/* 0x0250 */
+	__u64	async_enter_timer;		/* 0x0258 */
+	__u64	exit_timer;			/* 0x0260 */
+	__u64	user_timer;			/* 0x0268 */
+	__u64	system_timer;			/* 0x0270 */
+	__u64	steal_timer;			/* 0x0278 */
+	__u64	last_update_timer;		/* 0x0280 */
+	__u64	last_update_clock;		/* 0x0288 */
+
+	/* Current process. */
+	__u32	current_task;			/* 0x0290 */
+	__u32	thread_info;			/* 0x0294 */
+	__u32	kernel_stack;			/* 0x0298 */
+
+	/* Interrupt and panic stack. */
+	__u32	async_stack;			/* 0x029c */
+	__u32	panic_stack;			/* 0x02a0 */
+
+	/* Address space pointer. */
+	__u32	kernel_asce;			/* 0x02a4 */
+	__u32	user_asce;			/* 0x02a8 */
+	__u32	user_exec_asce;			/* 0x02ac */
+
+	/* SMP info area */
+	cpuid_t	cpu_id;				/* 0x02b0 */
+	__u32	cpu_nr;				/* 0x02b8 */
+	__u32	softirq_pending;		/* 0x02bc */
+	__u32	percpu_offset;			/* 0x02c0 */
+	__u32	ext_call_fast;			/* 0x02c4 */
+	__u64	int_clock;			/* 0x02c8 */
+	__u64	clock_comparator;		/* 0x02d0 */
+	__u8	pad_0x02d8[0x0300-0x02d8];	/* 0x02d8 */
+
+	/* Interrupt response block */
+	__u8	irb[64];			/* 0x0300 */
+
+	__u8	pad_0x0400[0x0e00-0x0400];	/* 0x0400 */
+
+	/*
+	 * 0xe00 contains the address of the IPL Parameter Information
+	 * block. Dump tools need IPIB for IPL after dump.
+	 * Note: do not change the position of any fields in 0x0e00-0x0f00
+	 */
+	__u32	ipib;				/* 0x0e00 */
+	__u32	ipib_checksum;			/* 0x0e04 */
+
+	/* Align to the top 1k of prefix area */
+	__u8	pad_0x0e08[0x1000-0x0e08];	/* 0x0e08 */
 #else /* !__s390x__ */
-        /* prefix area: defined by architecture */
-	__u32        ccw1[2];                  /* 0x000 */
-	__u32        ccw2[4];                  /* 0x008 */
-	__u8         pad1[0x80-0x18];          /* 0x018 */
-	__u32        ext_params;               /* 0x080 */
-	__u16        cpu_addr;                 /* 0x084 */
-	__u16        ext_int_code;             /* 0x086 */
-        __u16        svc_ilc;                  /* 0x088 */
-        __u16        svc_code;                 /* 0x08a */
-        __u16        pgm_ilc;                  /* 0x08c */
-        __u16        pgm_code;                 /* 0x08e */
-	__u32        data_exc_code;            /* 0x090 */
-	__u16        mon_class_num;            /* 0x094 */
-	__u16        per_perc_atmid;           /* 0x096 */
-	addr_t       per_address;              /* 0x098 */
-	__u8         exc_access_id;            /* 0x0a0 */
-	__u8         per_access_id;            /* 0x0a1 */
-	__u8         op_access_id;             /* 0x0a2 */
-	__u8         ar_access_id;             /* 0x0a3 */
-	__u8         pad2[0xA8-0xA4];          /* 0x0a4 */
-	addr_t	     trans_exc_code;	       /* 0x0a8 */
-	addr_t	     monitor_code;	       /* 0x0b0 */
-	__u16        subchannel_id;            /* 0x0b8 */
-	__u16        subchannel_nr;            /* 0x0ba */
-	__u32        io_int_parm;              /* 0x0bc */
-	__u32        io_int_word;              /* 0x0c0 */
-	__u8         pad3[0xc8-0xc4];          /* 0x0c4 */
-	__u32        stfl_fac_list;            /* 0x0c8 */
-	__u8         pad4[0xe8-0xcc];          /* 0x0cc */
-	__u32        mcck_interruption_code[2]; /* 0x0e8 */
-	__u8         pad5[0xf4-0xf0];          /* 0x0f0 */
-	__u32        external_damage_code;     /* 0x0f4 */
-	addr_t       failing_storage_address;  /* 0x0f8 */
-	__u8         pad6[0x120-0x100];        /* 0x100 */
-	psw_t        restart_old_psw;          /* 0x120 */
-	psw_t        external_old_psw;         /* 0x130 */
-	psw_t        svc_old_psw;              /* 0x140 */
-	psw_t        program_old_psw;          /* 0x150 */
-	psw_t        mcck_old_psw;             /* 0x160 */
-	psw_t        io_old_psw;               /* 0x170 */
-	__u8         pad7[0x1a0-0x180];        /* 0x180 */
-	psw_t        restart_psw;              /* 0x1a0 */
-	psw_t        external_new_psw;         /* 0x1b0 */
-	psw_t        svc_new_psw;              /* 0x1c0 */
-	psw_t        program_new_psw;          /* 0x1d0 */
-	psw_t        mcck_new_psw;             /* 0x1e0 */
-	psw_t        io_new_psw;               /* 0x1f0 */
-        psw_t        return_psw;               /* 0x200 */
-	__u8	     irb[64];		       /* 0x210 */
-	__u64        sync_enter_timer;         /* 0x250 */
-	__u64        async_enter_timer;        /* 0x258 */
-	__u64        exit_timer;               /* 0x260 */
-	__u64	     user_timer;	       /* 0x268 */
-	__u64	     system_timer;	       /* 0x270 */
-	__u64	     steal_timer;	       /* 0x278 */
-	__u64	     last_update_timer;        /* 0x280 */
-	__u64	     last_update_clock;        /* 0x288 */
-        psw_t        return_mcck_psw;          /* 0x290 */
-        __u8         pad8[0xc00-0x2a0];        /* 0x2a0 */
-        /* System info area */
-	__u64        save_area[16];            /* 0xc00 */
-        __u8         pad9[0xd40-0xc80];        /* 0xc80 */
- 	__u64        kernel_stack;             /* 0xd40 */
-	__u64        thread_info;              /* 0xd48 */
-	__u64        async_stack;              /* 0xd50 */
-	__u64        kernel_asce;              /* 0xd58 */
-	__u64        user_asce;                /* 0xd60 */
-	__u64        panic_stack;              /* 0xd68 */
-	__u64	     user_exec_asce;	       /* 0xd70 */
-	__u8	     pad10[0xd80-0xd78];       /* 0xd78 */
-	/* entry.S sensitive area start */
-	cpuid_t      cpu_id;		       /* 0xd80 */
-	__u32	     cpu_nr;		       /* 0xd88 */
-	__u8	     pad_0xd8c[0xdc0-0xd8c];   /* 0xd8c */
-	/* entry.S sensitive area end */
-
-        /* SMP info area: defined by DJB */
-	__u64	     clock_comparator;	       /* 0xdc0 */
-	__u64        ext_call_fast;            /* 0xdc8 */
-	__u64        percpu_offset;            /* 0xdd0 */
-	__u64        current_task;	       /* 0xdd8 */
-	__u32	     softirq_pending;	       /* 0xde0 */
-	__u32	     pad_0x0de4;	       /* 0xde4 */
-	__u64        int_clock;                /* 0xde8 */
-        __u8         pad12[0xe00-0xdf0];       /* 0xdf0 */
-
-	/* 0xe00 contains the address of the IPL Parameter */
-	/* Information block. Dump tools need IPIB for IPL */
-	/* after dump.					   */
-	__u64	     ipib;		       /* 0xe00 */
-	__u32	     ipib_checksum;	       /* 0xe08 */
+	/* 0x0000 - 0x01ff: defined by architecture */
+	__u32	ccw1[2];			/* 0x0000 */
+	__u32	ccw2[4];			/* 0x0008 */
+	__u8	pad_0x0018[0x0080-0x0018];	/* 0x0018 */
+	__u32	ext_params;			/* 0x0080 */
+	__u16	cpu_addr;			/* 0x0084 */
+	__u16	ext_int_code;			/* 0x0086 */
+	__u16	svc_ilc;			/* 0x0088 */
+	__u16	svc_code;			/* 0x008a */
+	__u16	pgm_ilc;			/* 0x008c */
+	__u16	pgm_code;			/* 0x008e */
+	__u32	data_exc_code;			/* 0x0090 */
+	__u16	mon_class_num;			/* 0x0094 */
+	__u16	per_perc_atmid;			/* 0x0096 */
+	addr_t	per_address;			/* 0x0098 */
+	__u8	exc_access_id;			/* 0x00a0 */
+	__u8	per_access_id;			/* 0x00a1 */
+	__u8	op_access_id;			/* 0x00a2 */
+	__u8	ar_access_id;			/* 0x00a3 */
+	__u8	pad_0x00a4[0x00a8-0x00a4];	/* 0x00a4 */
+	addr_t	trans_exc_code;			/* 0x00a8 */
+	addr_t	monitor_code;			/* 0x00b0 */
+	__u16	subchannel_id;			/* 0x00b8 */
+	__u16	subchannel_nr;			/* 0x00ba */
+	__u32	io_int_parm;			/* 0x00bc */
+	__u32	io_int_word;			/* 0x00c0 */
+	__u8	pad_0x00c4[0x00c8-0x00c4];	/* 0x00c4 */
+	__u32	stfl_fac_list;			/* 0x00c8 */
+	__u8	pad_0x00cc[0x00e8-0x00cc];	/* 0x00cc */
+	__u32	mcck_interruption_code[2];	/* 0x00e8 */
+	__u8	pad_0x00f0[0x00f4-0x00f0];	/* 0x00f0 */
+	__u32	external_damage_code;		/* 0x00f4 */
+	addr_t	failing_storage_address;	/* 0x00f8 */
+	__u8	pad_0x0100[0x0120-0x0100];	/* 0x0100 */
+	psw_t	restart_old_psw;		/* 0x0120 */
+	psw_t	external_old_psw;		/* 0x0130 */
+	psw_t	svc_old_psw;			/* 0x0140 */
+	psw_t	program_old_psw;		/* 0x0150 */
+	psw_t	mcck_old_psw;			/* 0x0160 */
+	psw_t	io_old_psw;			/* 0x0170 */
+	__u8	pad_0x0180[0x01a0-0x0180];	/* 0x0180 */
+	psw_t	restart_psw;			/* 0x01a0 */
+	psw_t	external_new_psw;		/* 0x01b0 */
+	psw_t	svc_new_psw;			/* 0x01c0 */
+	psw_t	program_new_psw;		/* 0x01d0 */
+	psw_t	mcck_new_psw;			/* 0x01e0 */
+	psw_t	io_new_psw;			/* 0x01f0 */
+
+	/* Entry/exit save area & return psws. */
+	__u64	save_area[16];			/* 0x0200 */
+	psw_t	return_psw;			/* 0x0280 */
+	psw_t	return_mcck_psw;		/* 0x0290 */
+
+	/* CPU accounting and timing values. */
+	__u64	sync_enter_timer;		/* 0x02a0 */
+	__u64	async_enter_timer;		/* 0x02a8 */
+	__u64	exit_timer;			/* 0x02b0 */
+	__u64	user_timer;			/* 0x02b8 */
+	__u64	system_timer;			/* 0x02c0 */
+	__u64	steal_timer;			/* 0x02c8 */
+	__u64	last_update_timer;		/* 0x02d0 */
+	__u64	last_update_clock;		/* 0x02d8 */
+
+	/* Current process. */
+	__u64	current_task;			/* 0x02e0 */
+	__u64	thread_info;			/* 0x02e8 */
+	__u64	kernel_stack;			/* 0x02f0 */
+
+	/* Interrupt and panic stack. */
+	__u64	async_stack;			/* 0x02f8 */
+	__u64	panic_stack;			/* 0x0300 */
+
+	/* Address space pointer. */
+	__u64	kernel_asce;			/* 0x0308 */
+	__u64	user_asce;			/* 0x0310 */
+	__u64	user_exec_asce;			/* 0x0318 */
+
+	/* SMP info area */
+	cpuid_t	cpu_id;				/* 0x0320 */
+	__u32	cpu_nr;				/* 0x0328 */
+	__u32	softirq_pending;		/* 0x032c */
+	__u64	percpu_offset;			/* 0x0330 */
+	__u64	ext_call_fast;			/* 0x0338 */
+	__u64	int_clock;			/* 0x0340 */
+	__u64	clock_comparator;		/* 0x0348 */
+	__u64	vdso_per_cpu_data;		/* 0x0350 */
+	__u8	pad_0x0358[0x0380-0x0358];	/* 0x0358 */
+
+	/* Interrupt response block. */
+	__u8	irb[64];			/* 0x0380 */
 
 	/* Per cpu primary space access list */
-	__u8	     pad_0xe0c[0xe38-0xe0c];   /* 0xe0c */
-	__u64	     vdso_per_cpu_data;	       /* 0xe38 */
-	__u32	     paste[16];		       /* 0xe40 */
-
-	__u8	     pad13[0x11b8-0xe80];      /* 0xe80 */
-
-	/* 64 bit extparam used for pfault, diag 250 etc  */
-	__u64        ext_params2;               /* 0x11B8 */
-
-	__u8         pad14[0x1200-0x11C0];      /* 0x11C0 */
-
-        /* System info area */ 
-
-	__u64        floating_pt_save_area[16]; /* 0x1200 */
-	__u64        gpregs_save_area[16];      /* 0x1280 */
-	__u32        st_status_fixed_logout[4]; /* 0x1300 */
-	__u8         pad15[0x1318-0x1310];      /* 0x1310 */
-	__u32        prefixreg_save_area;       /* 0x1318 */
-	__u32        fpt_creg_save_area;        /* 0x131c */
-	__u8         pad16[0x1324-0x1320];      /* 0x1320 */
-	__u32        tod_progreg_save_area;     /* 0x1324 */
-	__u32        cpu_timer_save_area[2];    /* 0x1328 */
-	__u32        clock_comp_save_area[2];   /* 0x1330 */
-	__u8         pad17[0x1340-0x1338];      /* 0x1338 */
-	__u32        access_regs_save_area[16]; /* 0x1340 */ 
-	__u64        cregs_save_area[16];       /* 0x1380 */
+	__u32	paste[16];			/* 0x03c0 */
+
+	__u8	pad_0x0400[0x0e00-0x0400];	/* 0x0400 */
+
+	/*
+	 * 0xe00 contains the address of the IPL Parameter Information
+	 * block. Dump tools need IPIB for IPL after dump.
+	 * Note: do not change the position of any fields in 0x0e00-0x0f00
+	 */
+	__u64	ipib;				/* 0x0e00 */
+	__u32	ipib_checksum;			/* 0x0e08 */
+	__u8	pad_0x0e0c[0x11b8-0x0e0c];	/* 0x0e0c */
+
+	/* 64 bit extparam used for pfault/diag 250: defined by architecture */
+	__u64	ext_params2;			/* 0x11B8 */
+	__u8	pad_0x11c0[0x1200-0x11C0];	/* 0x11C0 */
+
+	/* CPU register save area: defined by architecture */
+	__u64	floating_pt_save_area[16];	/* 0x1200 */
+	__u64	gpregs_save_area[16];		/* 0x1280 */
+	__u32	st_status_fixed_logout[4];	/* 0x1300 */
+	__u8	pad_0x1310[0x1318-0x1310];	/* 0x1310 */
+	__u32	prefixreg_save_area;		/* 0x1318 */
+	__u32	fpt_creg_save_area;		/* 0x131c */
+	__u8	pad_0x1320[0x1324-0x1320];	/* 0x1320 */
+	__u32	tod_progreg_save_area;		/* 0x1324 */
+	__u32	cpu_timer_save_area[2];		/* 0x1328 */
+	__u32	clock_comp_save_area[2];	/* 0x1330 */
+	__u8	pad_0x1338[0x1340-0x1338];	/* 0x1338 */
+	__u32	access_regs_save_area[16];	/* 0x1340 */
+	__u64	cregs_save_area[16];		/* 0x1380 */
 
 	/* align to the top of the prefix area */
-
-	__u8         pad18[0x2000-0x1400];      /* 0x1400 */
+	__u8	pad_0x1400[0x2000-0x1400];	/* 0x1400 */
 #endif /* !__s390x__ */
 } __attribute__((packed)); /* End structure*/
 
diff --git a/arch/s390/kernel/head.S b/arch/s390/kernel/head.S
index ec7e35f6055b..1046c2c9f8d1 100644
--- a/arch/s390/kernel/head.S
+++ b/arch/s390/kernel/head.S
@@ -469,6 +469,8 @@ start:
 	.org	0x10000
 startup:basr	%r13,0			# get base
 .LPG0:
+	xc	0x200(256),0x200	# partially clear lowcore
+	xc	0x300(256),0x300
 
 #ifndef CONFIG_MARCH_G5
 	# check processor version against MARCH_{G5,Z900,Z990,Z9_109,Z10}
-- 
cgit v1.2.3-55-g7522


From 159d1ff8f6c38086ed75f8e892790d0a4f3a6b71 Mon Sep 17 00:00:00 2001
From: Frank Munzert
Date: Thu, 26 Mar 2009 15:24:45 +0100
Subject: [S390] Use csum_partial in checksum.h

The cksm function in system.h is duplicate to csum_partial in checksum.h.
Remove cksm and use csum_partial instead.

Signed-off-by: Frank Munzert <munzert@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/system.h | 16 ----------------
 arch/s390/kernel/ipl.c         |  4 +++-
 drivers/s390/char/zcore.c      |  4 +++-
 3 files changed, 6 insertions(+), 18 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/system.h b/arch/s390/include/asm/system.h
index 3f2ccb82b863..3a8b26eb1f2e 100644
--- a/arch/s390/include/asm/system.h
+++ b/arch/s390/include/asm/system.h
@@ -458,22 +458,6 @@ static inline unsigned short stap(void)
 	return cpu_address;
 }
 
-static inline u32 cksm(void *addr, unsigned long len)
-{
-	register unsigned long _addr asm("0") = (unsigned long) addr;
-	register unsigned long _len asm("1") = len;
-	unsigned long accu = 0;
-
-	asm volatile(
-		"0:\n"
-		"	cksm	%0,%1\n"
-		"	jnz	0b\n"
-		: "+d" (accu), "+d" (_addr), "+d" (_len)
-		:
-		: "cc", "memory");
-	return accu;
-}
-
 extern void (*_machine_restart)(char *command);
 extern void (*_machine_halt)(void);
 extern void (*_machine_power_off)(void);
diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 5663c1f8e46a..505fec06e634 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -24,6 +24,7 @@
 #include <asm/reset.h>
 #include <asm/sclp.h>
 #include <asm/setup.h>
+#include <asm/checksum.h>
 
 #define IPL_PARM_BLOCK_VERSION 0
 
@@ -1359,7 +1360,8 @@ static void dump_reipl_run(struct shutdown_trigger *trigger)
 		"a" (&lowcore_ptr[smp_processor_id()]->ipib));
 #endif
 	asm volatile("stura %0,%1"
-		:: "a" (cksm(reipl_block_actual, reipl_block_actual->hdr.len)),
+		:: "a" (csum_partial(reipl_block_actual,
+				     reipl_block_actual->hdr.len, 0)),
 		"a" (&lowcore_ptr[smp_processor_id()]->ipib_checksum));
 	preempt_enable();
 	dump_run(trigger);
diff --git a/drivers/s390/char/zcore.c b/drivers/s390/char/zcore.c
index cfe782ee6473..1bbae433fbd8 100644
--- a/drivers/s390/char/zcore.c
+++ b/drivers/s390/char/zcore.c
@@ -24,6 +24,7 @@
 #include <asm/debug.h>
 #include <asm/processor.h>
 #include <asm/irqflags.h>
+#include <asm/checksum.h>
 #include "sclp.h"
 
 #define TRACE(x...) debug_sprintf_event(zcore_dbf, 1, x)
@@ -704,7 +705,8 @@ static int __init zcore_reipl_init(void)
 		free_page((unsigned long) ipl_block);
 		return rc;
 	}
-	if (cksm(ipl_block, ipl_block->hdr.len) != ipib_info.checksum) {
+	if (csum_partial(ipl_block, ipl_block->hdr.len, 0) !=
+	    ipib_info.checksum) {
 		TRACE("Checksum does not match\n");
 		free_page((unsigned long) ipl_block);
 		ipl_block = NULL;
-- 
cgit v1.2.3-55-g7522


From 59f2e69d0f95bc00353628ef33fd534fbb8e3597 Mon Sep 17 00:00:00 2001
From: Michael Holzheu
Date: Thu, 26 Mar 2009 15:24:46 +0100
Subject: [S390] zfcpdump: Prevent zcore from beeing built as a kernel module.

The zcore code switches to real addressing mode when creating a kernel dump.
This is not possible, if it is built as a kernel module. With this patch
zcore (zfcpdump) can't be built as a kernel module any more.

Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/Kconfig        | 2 +-
 arch/s390/kernel/setup.c | 4 ++--
 arch/s390/kernel/smp.c   | 4 ++--
 3 files changed, 5 insertions(+), 5 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 0a9463bea758..2a8af5e16345 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -545,7 +545,7 @@ config KEXEC
 	  but is independent of hardware/microcode support.
 
 config ZFCPDUMP
-	tristate "zfcpdump support"
+	bool "zfcpdump support"
 	select SMP
 	default n
 	help
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 91551ef1d67e..46fc981e02ba 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -226,7 +226,7 @@ static void __init conmode_default(void)
 	}
 }
 
-#if defined(CONFIG_ZFCPDUMP) || defined(CONFIG_ZFCPDUMP_MODULE)
+#ifdef CONFIG_ZFCPDUMP
 static void __init setup_zfcpdump(unsigned int console_devno)
 {
 	static char str[41];
@@ -515,7 +515,7 @@ static void __init setup_memory_end(void)
 	unsigned long max_mem;
 	int i;
 
-#if defined(CONFIG_ZFCPDUMP) || defined(CONFIG_ZFCPDUMP_MODULE)
+#ifdef CONFIG_ZFCPDUMP
 	if (ipl_info.type == IPL_TYPE_FCP_DUMP) {
 		memory_end = ZFCPDUMP_HSA_SIZE;
 		memory_end_set = 1;
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 775885772ff1..bb3d34aa7b85 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -226,7 +226,7 @@ EXPORT_SYMBOL(smp_ctl_clear_bit);
  */
 #define CPU_INIT_NO	1
 
-#if defined(CONFIG_ZFCPDUMP) || defined(CONFIG_ZFCPDUMP_MODULE)
+#ifdef CONFIG_ZFCPDUMP
 
 /*
  * zfcpdump_prefix_array holds prefix registers for the following scenario:
@@ -267,7 +267,7 @@ EXPORT_SYMBOL_GPL(zfcpdump_save_areas);
 
 static inline void smp_get_save_area(unsigned int cpu, unsigned int phy_cpu) { }
 
-#endif /* CONFIG_ZFCPDUMP || CONFIG_ZFCPDUMP_MODULE */
+#endif /* CONFIG_ZFCPDUMP */
 
 static int cpu_stopped(int cpu)
 {
-- 
cgit v1.2.3-55-g7522


From 6aa0d3a922c4f58fc36cc1502c6ac72f999e26bb Mon Sep 17 00:00:00 2001
From: Stoyan Gaydarov
Date: Thu, 26 Mar 2009 15:24:47 +0100
Subject: [S390] BUG to BUG_ON changes

Signed-off-by: Stoyan Gaydarov <stoyboyker@gmail.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/debug.c      | 3 +--
 arch/s390/kernel/setup.c      | 3 +--
 drivers/s390/char/tape_3590.c | 3 +--
 drivers/s390/char/tape_core.c | 6 ++----
 drivers/s390/char/tape_std.c  | 4 ++--
 5 files changed, 7 insertions(+), 12 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index ba03fc0a3a56..39137b9e1622 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -698,8 +698,7 @@ debug_info_t *debug_register_mode(const char *name, int pages_per_area,
 	if ((uid != 0) || (gid != 0))
 		pr_warning("Root becomes the owner of all s390dbf files "
 			   "in sysfs\n");
-	if (!initialized)
-		BUG();
+	BUG_ON(!initialized);
 	mutex_lock(&debug_mutex);
 
         /* create new debug_info */
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 46fc981e02ba..580abb53ce83 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -134,8 +134,7 @@ void __cpuinit cpu_init(void)
 
 	atomic_inc(&init_mm.mm_count);
 	current->active_mm = &init_mm;
-        if (current->mm)
-                BUG();
+	BUG_ON(current->mm);
         enter_lazy_tlb(&init_mm, current);
 }
 
diff --git a/drivers/s390/char/tape_3590.c b/drivers/s390/char/tape_3590.c
index 5a5bb97a081a..fc1d91294143 100644
--- a/drivers/s390/char/tape_3590.c
+++ b/drivers/s390/char/tape_3590.c
@@ -664,8 +664,7 @@ tape_3590_bread(struct tape_device *device, struct request *req)
 			ccw++;
 			dst += TAPEBLOCK_HSEC_SIZE;
 		}
-		if (off > bv->bv_len)
-			BUG();
+		BUG_ON(off > bv->bv_len);
 	}
 	ccw = tape_ccw_end(ccw, NOP, 0, NULL);
 	DBF_EVENT(6, "xBREDccwg\n");
diff --git a/drivers/s390/char/tape_core.c b/drivers/s390/char/tape_core.c
index 1b6a24412465..08c09d3503cf 100644
--- a/drivers/s390/char/tape_core.c
+++ b/drivers/s390/char/tape_core.c
@@ -624,8 +624,7 @@ tape_alloc_request(int cplength, int datasize)
 {
 	struct tape_request *request;
 
-	if (datasize > PAGE_SIZE || (cplength*sizeof(struct ccw1)) > PAGE_SIZE)
-		BUG();
+	BUG_ON(datasize > PAGE_SIZE || (cplength*sizeof(struct ccw1)) > PAGE_SIZE);
 
 	DBF_LH(6, "tape_alloc_request(%d, %d)\n", cplength, datasize);
 
@@ -782,8 +781,7 @@ static void tape_long_busy_timeout(unsigned long data)
 	device = (struct tape_device *) data;
 	spin_lock_irq(get_ccwdev_lock(device->cdev));
 	request = list_entry(device->req_queue.next, struct tape_request, list);
-	if (request->status != TAPE_REQUEST_LONG_BUSY)
-		BUG();
+	BUG_ON(request->status != TAPE_REQUEST_LONG_BUSY);
 	DBF_LH(6, "%08x: Long busy timeout.\n", device->cdev_id);
 	__tape_start_next_request(device);
 	device->lb_timeout.data = (unsigned long) tape_put_device(device);
diff --git a/drivers/s390/char/tape_std.c b/drivers/s390/char/tape_std.c
index 44dd0f80c847..1a9420ba518d 100644
--- a/drivers/s390/char/tape_std.c
+++ b/drivers/s390/char/tape_std.c
@@ -37,8 +37,8 @@ tape_std_assign_timeout(unsigned long data)
 	int rc;
 
 	request = (struct tape_request *) data;
-	if ((device = request->device) == NULL)
-		BUG();
+	device = request->device;
+	BUG_ON(!device);
 
 	DBF_EVENT(3, "%08x: Assignment timeout. Device busy.\n",
 			device->cdev_id);
-- 
cgit v1.2.3-55-g7522


From 3e75a902196c45d26d5e28014eb2d9821aa9794f Mon Sep 17 00:00:00 2001
From: Johannes Weiner
Date: Thu, 26 Mar 2009 15:24:48 +0100
Subject: [S390] use kzfree()

Use kzfree() instead of memset() + kfree().

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/crypto/prng.c             | 3 +--
 drivers/s390/crypto/zcrypt_pcixcc.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/crypto/prng.c b/arch/s390/crypto/prng.c
index eca724d229ec..b49c00ce65e9 100644
--- a/arch/s390/crypto/prng.c
+++ b/arch/s390/crypto/prng.c
@@ -201,8 +201,7 @@ out_free:
 static void __exit prng_exit(void)
 {
 	/* wipe me */
-	memset(p->buf, 0, prng_chunk_size);
-	kfree(p->buf);
+	kzfree(p->buf);
 	kfree(p);
 
 	misc_deregister(&prng_dev);
diff --git a/drivers/s390/crypto/zcrypt_pcixcc.c b/drivers/s390/crypto/zcrypt_pcixcc.c
index e7a1e22e77ac..c20d4790258e 100644
--- a/drivers/s390/crypto/zcrypt_pcixcc.c
+++ b/drivers/s390/crypto/zcrypt_pcixcc.c
@@ -781,8 +781,7 @@ static long zcrypt_pcixcc_send_cprb(struct zcrypt_device *zdev,
 		/* Signal pending. */
 		ap_cancel_message(zdev->ap_dev, &ap_msg);
 out_free:
-	memset(ap_msg.message, 0x0, ap_msg.length);
-	kfree(ap_msg.message);
+	kzfree(ap_msg.message);
 	return rc;
 }
 
-- 
cgit v1.2.3-55-g7522


From a0fa8e3743c80643a06a770b95e73e751548ab17 Mon Sep 17 00:00:00 2001
From: Michael Holzheu
Date: Thu, 26 Mar 2009 15:24:49 +0100
Subject: [S390] s390dbf: Remove redundant initilizations.

Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/debug.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index 39137b9e1622..369de12b8733 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -603,7 +603,7 @@ debug_input(struct file *file, const char __user *user_buf, size_t length,
 static int
 debug_open(struct inode *inode, struct file *file)
 {
-	int i = 0, rc = 0;
+	int i, rc = 0;
 	file_private_info_t *p_info;
 	debug_info_t *debug_info, *debug_info_snapshot;
 
@@ -1155,7 +1155,6 @@ debug_unregister_view(debug_info_t * id, struct debug_view *view)
 	else {
 		debugfs_remove(id->debugfs_entries[i]);
 		id->views[i] = NULL;
-		rc = 0;
 	}
 	spin_unlock_irqrestore(&id->lock, flags);
 out:
-- 
cgit v1.2.3-55-g7522


From 58ace9f2a8f1caa0303aa64079d3929be28a937a Mon Sep 17 00:00:00 2001
From: Michael Holzheu
Date: Thu, 26 Mar 2009 15:24:50 +0100
Subject: [S390] s390dbf: Remove needless check for NULL pointer.

The check for NULL pointer has already be done before. Therefore we can remove
the second check.

Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/debug.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index 369de12b8733..be8bceaf37d9 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -642,8 +642,7 @@ found:
 	p_info = kmalloc(sizeof(file_private_info_t),
 						GFP_KERNEL);
 	if(!p_info){
-		if(debug_info_snapshot)
-			debug_info_free(debug_info_snapshot);
+		debug_info_free(debug_info_snapshot);
 		rc = -ENOMEM;
 		goto out;
 	}
-- 
cgit v1.2.3-55-g7522


From 008d2d112cb8a743a87dfe41a67e11c0a4c73aa4 Mon Sep 17 00:00:00 2001
From: Michael Holzheu
Date: Thu, 26 Mar 2009 15:24:51 +0100
Subject: [S390] ipl: Improve checking logic and remove switch defaults.

A code analysis tool reported two warnings:
"The expression `ipl_info.type == IPL_TYPE_FCP' is true whenever evaluated."
and "Default is not possible". This patch improves the corresponding if
statement logic and removes the unnecessary switch defaults.

Signed-off-by:  Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/ipl.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 505fec06e634..2b901fb8193b 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -966,7 +966,6 @@ static void reipl_run(struct shutdown_trigger *trigger)
 		diag308(DIAG308_IPL, NULL);
 		break;
 	case REIPL_METHOD_FCP_DUMP:
-	default:
 		break;
 	}
 	disabled_wait((unsigned long) __builtin_return_address(0));
@@ -1075,10 +1074,12 @@ static int __init reipl_fcp_init(void)
 {
 	int rc;
 
-	if ((!diag308_set_works) && (ipl_info.type != IPL_TYPE_FCP))
-		return 0;
-	if ((!diag308_set_works) && (ipl_info.type == IPL_TYPE_FCP))
-		make_attrs_ro(reipl_fcp_attrs);
+	if (!diag308_set_works) {
+		if (ipl_info.type == IPL_TYPE_FCP)
+			make_attrs_ro(reipl_fcp_attrs);
+		else
+			return 0;
+	}
 
 	reipl_block_fcp = (void *) get_zeroed_page(GFP_KERNEL);
 	if (!reipl_block_fcp)
@@ -1259,7 +1260,6 @@ static void dump_run(struct shutdown_trigger *trigger)
 		diag308(DIAG308_DUMP, NULL);
 		break;
 	case DUMP_METHOD_NONE:
-	default:
 		return;
 	}
 	printk(KERN_EMERG "Dump failed!\n");
@@ -1746,7 +1746,6 @@ void __init setup_ipl(void)
 			sizeof(ipl_info.data.nss.name));
 		break;
 	case IPL_TYPE_UNKNOWN:
-	default:
 		/* We have no info to copy */
 		break;
 	}
-- 
cgit v1.2.3-55-g7522


From 6d54c5a3fb13d32d66a8383cb0b5fb422a563051 Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:52 +0100
Subject: [S390] smp: fix memory leak on __cpu_up

If sigp_set_prefix fails on __cpu_up we leak the lowcore structures
and async+panic stacks for the failed cpu.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/smp.c  | 6 +++---
 arch/s390/kernel/vdso.c | 2 --
 2 files changed, 3 insertions(+), 5 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index bb3d34aa7b85..1f4711b60aab 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -508,7 +508,6 @@ out:
 	return -ENOMEM;
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
 static void smp_free_lowcore(int cpu)
 {
 	struct _lowcore *lowcore;
@@ -527,7 +526,6 @@ static void smp_free_lowcore(int cpu)
 	free_pages((unsigned long) lowcore, lc_order);
 	lowcore_ptr[cpu] = NULL;
 }
-#endif /* CONFIG_HOTPLUG_CPU */
 
 /* Upping and downing of CPUs */
 int __cpuinit __cpu_up(unsigned int cpu)
@@ -544,8 +542,10 @@ int __cpuinit __cpu_up(unsigned int cpu)
 
 	ccode = signal_processor_p((__u32)(unsigned long)(lowcore_ptr[cpu]),
 				   cpu, sigp_set_prefix);
-	if (ccode)
+	if (ccode) {
+		smp_free_lowcore(cpu);
 		return -EIO;
+	}
 
 	idle = current_set[cpu];
 	cpu_lowcore = lowcore_ptr[cpu];
diff --git a/arch/s390/kernel/vdso.c b/arch/s390/kernel/vdso.c
index 690e17819686..89b2e7f1b7a9 100644
--- a/arch/s390/kernel/vdso.c
+++ b/arch/s390/kernel/vdso.c
@@ -144,7 +144,6 @@ out:
 	return -ENOMEM;
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
 void vdso_free_per_cpu(int cpu, struct _lowcore *lowcore)
 {
 	unsigned long segment_table, page_table, page_frame;
@@ -163,7 +162,6 @@ void vdso_free_per_cpu(int cpu, struct _lowcore *lowcore)
 	free_page(page_table);
 	free_pages(segment_table, SEGMENT_ORDER);
 }
-#endif /* CONFIG_HOTPLUG_CPU */
 
 static void __vdso_init_cr5(void *dummy)
 {
-- 
cgit v1.2.3-55-g7522


From d0d3cdf4c27fa4ce241616da08138954e02890f7 Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:53 +0100
Subject: [S390] smp: perform initial cpu reset before starting a cpu

Performing an initial cpu reset makes sure all registers and tlbs of
the targeted cpu are initialized and flushed.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/smp.c | 23 ++++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 1f4711b60aab..fe5f8dc1e6fa 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -534,18 +534,23 @@ int __cpuinit __cpu_up(unsigned int cpu)
 	struct _lowcore *cpu_lowcore;
 	struct stack_frame *sf;
 	sigp_ccode ccode;
+	u32 lowcore;
 
 	if (smp_cpu_state[cpu] != CPU_STATE_CONFIGURED)
 		return -EIO;
 	if (smp_alloc_lowcore(cpu))
 		return -ENOMEM;
-
-	ccode = signal_processor_p((__u32)(unsigned long)(lowcore_ptr[cpu]),
-				   cpu, sigp_set_prefix);
-	if (ccode) {
-		smp_free_lowcore(cpu);
-		return -EIO;
-	}
+	do {
+		ccode = signal_processor(cpu, sigp_initial_cpu_reset);
+		if (ccode == sigp_busy)
+			udelay(10);
+		if (ccode == sigp_not_operational)
+			goto err_out;
+	} while (ccode == sigp_busy);
+
+	lowcore = (u32)(unsigned long)lowcore_ptr[cpu];
+	while (signal_processor_p(lowcore, cpu, sigp_set_prefix) == sigp_busy)
+		udelay(10);
 
 	idle = current_set[cpu];
 	cpu_lowcore = lowcore_ptr[cpu];
@@ -574,6 +579,10 @@ int __cpuinit __cpu_up(unsigned int cpu)
 	while (!cpu_online(cpu))
 		cpu_relax();
 	return 0;
+
+err_out:
+	smp_free_lowcore(cpu);
+	return -EIO;
 }
 
 static int __init setup_possible_cpus(char *s)
-- 
cgit v1.2.3-55-g7522


From 2ac3307f275c2a91af0417e16d2cfb95ae478661 Mon Sep 17 00:00:00 2001
From: Christian Ehrhardt
Date: Thu, 26 Mar 2009 15:24:54 +0100
Subject: [S390] fix dfp elf hwcap/facility bit detection

The old dfp detection wanted to check bit 43 (dfp high performance), but due
to a wrong calculation always used to check bit 42. Additionally the
"userspace expectation" is, that the dfp capability bit is set is if facility
bit 42 (decimal floating point facility available) and bit 44 (perform floating
point operation facility avail).
The patch fixes the bit calculation and extends the check to work like:
  elf hw cap dfp bit = facility bits 42 (dfp) & 44 (pfpo) available

Signed-off-by: Christian Ehrhardt <ehrhardt@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/setup.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 580abb53ce83..18222ac4078f 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -713,13 +713,15 @@ static void __init setup_hwcaps(void)
 	 * How many facility words are stored depends on the number of
 	 * doublewords passed to the instruction. The additional facilites
 	 * are:
-	 *   Bit 43: decimal floating point facility is installed
+	 *   Bit 42: decimal floating point facility is installed
+	 *   Bit 44: perform floating point operation facility is installed
 	 * translated to:
 	 *   HWCAP_S390_DFP bit 6.
 	 */
 	if ((elf_hwcap & (1UL << 2)) &&
 	    __stfle(&facility_list_extended, 1) > 0) {
-		if (facility_list_extended & (1ULL << (64 - 43)))
+		if ((facility_list_extended & (1ULL << (63 - 42)))
+		   && (facility_list_extended & (1ULL << (63 - 44))))
 			elf_hwcap |= 1UL << 6;
 	}
 
-- 
cgit v1.2.3-55-g7522


From 7e9b580e5f0644cd8952b6671fd5380fd430bca3 Mon Sep 17 00:00:00 2001
From: Michael Holzheu
Date: Thu, 26 Mar 2009 15:24:55 +0100
Subject: [S390] Ensure that ipl panic notifier is called late.

The s390 ipl panic notifier will stop the system or trigger a system dump.
This should be done as final action on the panic path. All other panic
notifiers should be executed before. Currently we use priority 0 for the ipl
notifier. In order to be called late, this patch changes the priority to
INT_MIN which is the lowest possible priority.

Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/ipl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 2b901fb8193b..56d876904ecc 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -1722,7 +1722,7 @@ static int on_panic_notify(struct notifier_block *self,
 
 static struct notifier_block on_panic_nb = {
 	.notifier_call = on_panic_notify,
-	.priority = 0,
+	.priority = INT_MIN,
 };
 
 void __init setup_ipl(void)
-- 
cgit v1.2.3-55-g7522


From 488253ce49714f4e9d42413c1d60b7724059a338 Mon Sep 17 00:00:00 2001
From: Andreas Krebbel
Date: Thu, 26 Mar 2009 15:24:56 +0100
Subject: [S390] Add hwcap flag for the etf3 enhancement facility

The Extended Translation Facility 3 (ETF3) added instructions which
allow conversions between different unicode character maps (UTF-8,
UTF-32 ...).  These instructions got enhanced with a later version of
the ETF3 allowing malformed multibyte chars to be recognized and
reported correctly.  The attached patch reserves bit 8 in the elf
hwcaps vector for the enhanced version of ETF3. The bit corresponds to
the stfle bits 22 and 30 and will only be set if both of the stfle
bits are set.

Signed-off-by: Andreas Krebbel <krebbel@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/processor.c |  6 +++---
 arch/s390/kernel/setup.c     | 13 ++++++++++---
 2 files changed, 13 insertions(+), 6 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/processor.c b/arch/s390/kernel/processor.c
index 423da1bd42a4..802c8ab247f3 100644
--- a/arch/s390/kernel/processor.c
+++ b/arch/s390/kernel/processor.c
@@ -31,9 +31,9 @@ void __cpuinit print_cpu_info(void)
 
 static int show_cpuinfo(struct seq_file *m, void *v)
 {
-	static const char *hwcap_str[8] = {
+	static const char *hwcap_str[9] = {
 		"esan3", "zarch", "stfle", "msa", "ldisp", "eimm", "dfp",
-		"edat"
+		"edat", "etf3eh"
 	};
 	struct _lowcore *lc;
 	unsigned long n = (unsigned long) v - 1;
@@ -48,7 +48,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 			   num_online_cpus(), loops_per_jiffy/(500000/HZ),
 			   (loops_per_jiffy/(5000/HZ))%100);
 		seq_puts(m, "features\t: ");
-		for (i = 0; i < 8; i++)
+		for (i = 0; i < 9; i++)
 			if (hwcap_str[i] && (elf_hwcap & (1UL << i)))
 				seq_printf(m, "%s ", hwcap_str[i]);
 		seq_puts(m, "\n");
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index 18222ac4078f..06201b93cbbf 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -696,15 +696,22 @@ static void __init setup_hwcaps(void)
 	 *   Bit 17: the message-security assist is installed
 	 *   Bit 19: the long-displacement facility is installed
 	 *   Bit 21: the extended-immediate facility is installed
+	 *   Bit 22: extended-translation facility 3 is installed
+	 *   Bit 30: extended-translation facility 3 enhancement facility
 	 * These get translated to:
 	 *   HWCAP_S390_ESAN3 bit 0, HWCAP_S390_ZARCH bit 1,
 	 *   HWCAP_S390_STFLE bit 2, HWCAP_S390_MSA bit 3,
-	 *   HWCAP_S390_LDISP bit 4, and HWCAP_S390_EIMM bit 5.
+	 *   HWCAP_S390_LDISP bit 4, HWCAP_S390_EIMM bit 5 and
+	 *   HWCAP_S390_ETF3EH bit 8 (22 && 30).
 	 */
 	for (i = 0; i < 6; i++)
 		if (facility_list & (1UL << (31 - stfl_bits[i])))
 			elf_hwcap |= 1UL << i;
 
+	if ((facility_list & (1UL << (31 - 22)))
+	    && (facility_list & (1UL << (31 - 30))))
+		elf_hwcap |= 1UL << 8;
+
 	/*
 	 * Check for additional facilities with store-facility-list-extended.
 	 * stfle stores doublewords (8 byte) with bit 1ULL<<63 as bit 0
@@ -716,12 +723,12 @@ static void __init setup_hwcaps(void)
 	 *   Bit 42: decimal floating point facility is installed
 	 *   Bit 44: perform floating point operation facility is installed
 	 * translated to:
-	 *   HWCAP_S390_DFP bit 6.
+	 *   HWCAP_S390_DFP bit 6 (42 && 44).
 	 */
 	if ((elf_hwcap & (1UL << 2)) &&
 	    __stfle(&facility_list_extended, 1) > 0) {
 		if ((facility_list_extended & (1ULL << (63 - 42)))
-		   && (facility_list_extended & (1ULL << (63 - 44))))
+		    && (facility_list_extended & (1ULL << (63 - 44))))
 			elf_hwcap |= 1UL << 6;
 	}
 
-- 
cgit v1.2.3-55-g7522


From c70d0fef1da286d7f96e774674aea2d20cf6222f Mon Sep 17 00:00:00 2001
From: Heiko Carstens
Date: Thu, 26 Mar 2009 15:24:57 +0100
Subject: [S390] fix clock comparator save area usage

The lowcore clock comparator save area on 64 bit machines is defined to
contain only the seven most significant bits of the register.
That's also why it starts at an uneven address (0x1331).
The current code however writes eight bytes to the address and
therefore overwrites the first byte of the access register save area.
Fix this and write only seven bytes to the save area.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/reipl64.S | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/reipl64.S b/arch/s390/kernel/reipl64.S
index c41930499a5f..774147824c3d 100644
--- a/arch/s390/kernel/reipl64.S
+++ b/arch/s390/kernel/reipl64.S
@@ -1,10 +1,7 @@
 /*
- *  arch/s390/kernel/reipl.S
- *
- *  S390 version
- *    Copyright (C) 2000 IBM Deutschland Entwicklung GmbH, IBM Corporation
- *    Author(s): Holger Smolinski (Holger.Smolinski@de.ibm.com)
-		 Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com)
+ *    Copyright IBM Corp 2000,2009
+ *    Author(s): Holger Smolinski <Holger.Smolinski@de.ibm.com>,
+ *		 Denis Joseph Barrow,
  */
 
 #include <asm/lowcore.h>
@@ -30,7 +27,7 @@ do_reipl_asm:	basr	%r13,0
 		mvc	__LC_PREFIX_SAVE_AREA-0x1000(4,%r1),0(%r10)
 		stfpc	__LC_FP_CREG_SAVE_AREA-0x1000(%r1)
 		stckc	.Lclkcmp-.Lpg0(%r13)
-		mvc	__LC_CLOCK_COMP_SAVE_AREA-0x1000(8,%r1),.Lclkcmp-.Lpg0(%r13)
+		mvc	__LC_CLOCK_COMP_SAVE_AREA-0x1000(7,%r1),.Lclkcmp-.Lpg0(%r13)
 		stpt	__LC_CPU_TIMER_SAVE_AREA-0x1000(%r1)
 		stg	%r13, __LC_PSW_SAVE_AREA-0x1000+8(%r1)
 
-- 
cgit v1.2.3-55-g7522


From 6f7a321d5feb0bc9aa7f7b14eceb1e6a63bff937 Mon Sep 17 00:00:00 2001
From: Rusty Russell
Date: Thu, 26 Mar 2009 15:24:58 +0100
Subject: [S390] cpumask: remove cpu_coregroup_map

Impact: cleanup

cpu_coregroup_mask is the New Hotness.

As S/390 uses theirs internally, so we just make it static.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/topology.h | 1 -
 arch/s390/kernel/topology.c      | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h
index c979c3b56ab0..5e0ad618dc45 100644
--- a/arch/s390/include/asm/topology.h
+++ b/arch/s390/include/asm/topology.h
@@ -5,7 +5,6 @@
 
 #define mc_capable()	(1)
 
-cpumask_t cpu_coregroup_map(unsigned int cpu);
 const struct cpumask *cpu_coregroup_mask(unsigned int cpu);
 
 extern cpumask_t cpu_core_map[NR_CPUS];
diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index cc362c9ea8f1..3c72c9cf22b6 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -74,7 +74,7 @@ static DEFINE_SPINLOCK(topology_lock);
 
 cpumask_t cpu_core_map[NR_CPUS];
 
-cpumask_t cpu_coregroup_map(unsigned int cpu)
+static cpumask_t cpu_coregroup_map(unsigned int cpu)
 {
 	struct core_info *core = &core_info;
 	unsigned long flags;
-- 
cgit v1.2.3-55-g7522


From 93632d1bf7447d445c6fc274c8931152f9412b56 Mon Sep 17 00:00:00 2001
From: Rusty Russell
Date: Thu, 26 Mar 2009 15:24:59 +0100
Subject: [S390] cpumask: prepare for iterators to only go to
 nr_cpu_ids/nr_cpumask_bits.

Impact: cleanup, futureproof

In fact, all cpumask ops will only be valid (in general) for bit
numbers < nr_cpu_ids.  So use that instead of NR_CPUS in various
places (I also updated the immediate sites to use the new cpumask_
operators).

This is always safe: no cpu number can be >= nr_cpu_ids, and
nr_cpu_ids is initialized to NR_CPUS at boot.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/smp.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index fe5f8dc1e6fa..13052a4ed71d 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -297,8 +297,8 @@ static int smp_rescan_cpus_sigp(cpumask_t avail)
 {
 	int cpu_id, logical_cpu;
 
-	logical_cpu = first_cpu(avail);
-	if (logical_cpu == NR_CPUS)
+	logical_cpu = cpumask_first(&avail);
+	if (logical_cpu >= nr_cpu_ids)
 		return 0;
 	for (cpu_id = 0; cpu_id <= 65535; cpu_id++) {
 		if (cpu_known(cpu_id))
@@ -309,8 +309,8 @@ static int smp_rescan_cpus_sigp(cpumask_t avail)
 			continue;
 		cpu_set(logical_cpu, cpu_present_map);
 		smp_cpu_state[logical_cpu] = CPU_STATE_CONFIGURED;
-		logical_cpu = next_cpu(logical_cpu, avail);
-		if (logical_cpu == NR_CPUS)
+		logical_cpu = cpumask_next(logical_cpu, &avail);
+		if (logical_cpu >= nr_cpu_ids)
 			break;
 	}
 	return 0;
@@ -322,8 +322,8 @@ static int smp_rescan_cpus_sclp(cpumask_t avail)
 	int cpu_id, logical_cpu, cpu;
 	int rc;
 
-	logical_cpu = first_cpu(avail);
-	if (logical_cpu == NR_CPUS)
+	logical_cpu = cpumask_first(&avail);
+	if (logical_cpu >= nr_cpu_ids)
 		return 0;
 	info = kmalloc(sizeof(*info), GFP_KERNEL);
 	if (!info)
@@ -344,8 +344,8 @@ static int smp_rescan_cpus_sclp(cpumask_t avail)
 			smp_cpu_state[logical_cpu] = CPU_STATE_STANDBY;
 		else
 			smp_cpu_state[logical_cpu] = CPU_STATE_CONFIGURED;
-		logical_cpu = next_cpu(logical_cpu, avail);
-		if (logical_cpu == NR_CPUS)
+		logical_cpu = cpumask_next(logical_cpu, &avail);
+		if (logical_cpu >= nr_cpu_ids)
 			break;
 	}
 out:
@@ -591,7 +591,7 @@ static int __init setup_possible_cpus(char *s)
 
 	pcpus = simple_strtoul(s, NULL, 0);
 	cpu_possible_map = cpumask_of_cpu(0);
-	for (cpu = 1; cpu < pcpus && cpu < NR_CPUS; cpu++)
+	for (cpu = 1; cpu < pcpus && cpu < nr_cpu_ids; cpu++)
 		cpu_set(cpu, cpu_possible_map);
 	return 0;
 }
-- 
cgit v1.2.3-55-g7522


From def6cfb70bab83c0094bc0cedd27c4eda563043e Mon Sep 17 00:00:00 2001
From: Rusty Russell
Date: Thu, 26 Mar 2009 15:25:00 +0100
Subject: [S390] cpumask: Use accessors code.

Impact: use new API

Use the accessors rather than frobbing bits directly.  Most of this is
in arch code I haven't even compiled, but is straightforward.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/smp.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 13052a4ed71d..006ed5016eb4 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -590,9 +590,8 @@ static int __init setup_possible_cpus(char *s)
 	int pcpus, cpu;
 
 	pcpus = simple_strtoul(s, NULL, 0);
-	cpu_possible_map = cpumask_of_cpu(0);
-	for (cpu = 1; cpu < pcpus && cpu < nr_cpu_ids; cpu++)
-		cpu_set(cpu, cpu_possible_map);
+	for (cpu = 0; cpu < pcpus && cpu < nr_cpu_ids; cpu++)
+		set_cpu_possible(cpu, true);
 	return 0;
 }
 early_param("possible_cpus", setup_possible_cpus);
-- 
cgit v1.2.3-55-g7522


From 005f8eee6f3c8173e492d7bd4d51bda990eb468b Mon Sep 17 00:00:00 2001
From: Rusty Russell
Date: Thu, 26 Mar 2009 15:25:01 +0100
Subject: [S390] cpumask: use mm_cpumask() wrapper

Makes code futureproof against the impending change to mm->cpu_vm_mask.

It's also a chance to use the new cpumask_ ops which take a pointer
(the older ones are deprecated, but there's no hurry for arch code).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/include/asm/mmu_context.h | 2 +-
 arch/s390/include/asm/tlbflush.h    | 4 ++--
 arch/s390/mm/pgtable.c              | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index 28ec870655af..fc7edd6f41b6 100644
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -74,7 +74,7 @@ static inline void update_mm(struct mm_struct *mm, struct task_struct *tsk)
 static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 			     struct task_struct *tsk)
 {
-	cpu_set(smp_processor_id(), next->cpu_vm_mask);
+	cpumask_set_cpu(smp_processor_id(), mm_cpumask(next));
 	update_mm(next, tsk);
 }
 
diff --git a/arch/s390/include/asm/tlbflush.h b/arch/s390/include/asm/tlbflush.h
index d60394b9745e..304cffa623e1 100644
--- a/arch/s390/include/asm/tlbflush.h
+++ b/arch/s390/include/asm/tlbflush.h
@@ -51,7 +51,7 @@ static inline void __tlb_flush_full(struct mm_struct *mm)
 	 * If the process only ran on the local cpu, do a local flush.
 	 */
 	local_cpumask = cpumask_of_cpu(smp_processor_id());
-	if (cpus_equal(mm->cpu_vm_mask, local_cpumask))
+	if (cpumask_equal(mm_cpumask(mm), &local_cpumask))
 		__tlb_flush_local();
 	else
 		__tlb_flush_global();
@@ -73,7 +73,7 @@ static inline void __tlb_flush_idte(unsigned long asce)
 
 static inline void __tlb_flush_mm(struct mm_struct * mm)
 {
-	if (unlikely(cpus_empty(mm->cpu_vm_mask)))
+	if (unlikely(cpumask_empty(mm_cpumask(mm))))
 		return;
 	/*
 	 * If the machine has IDTE we prefer to do a per mm flush
diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c
index 9bf86125f6f3..be6c1cf4ad5a 100644
--- a/arch/s390/mm/pgtable.c
+++ b/arch/s390/mm/pgtable.c
@@ -296,7 +296,7 @@ int s390_enable_sie(void)
 	tsk->mm = tsk->active_mm = mm;
 	preempt_disable();
 	update_mm(mm, tsk);
-	cpu_set(smp_processor_id(), mm->cpu_vm_mask);
+	cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm));
 	preempt_enable();
 	task_unlock(tsk);
 	mmput(old_mm);
-- 
cgit v1.2.3-55-g7522


From ef3500b2b2955af4fa6b0564b51c0c604e38c571 Mon Sep 17 00:00:00 2001
From: Huang Weiyi
Date: Thu, 26 Mar 2009 15:25:02 +0100
Subject: [S390] remove duplicated #includes

Remove duplicated #include's in arch/s390/kernel/ipl.c.

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
---
 arch/s390/kernel/ipl.c | 1 -
 1 file changed, 1 deletion(-)

(limited to 'arch/s390')

diff --git a/arch/s390/kernel/ipl.c b/arch/s390/kernel/ipl.c
index 56d876904ecc..6f3711a0eaaa 100644
--- a/arch/s390/kernel/ipl.c
+++ b/arch/s390/kernel/ipl.c
@@ -23,7 +23,6 @@
 #include <asm/ebcdic.h>
 #include <asm/reset.h>
 #include <asm/sclp.h>
-#include <asm/setup.h>
 #include <asm/checksum.h>
 
 #define IPL_PARM_BLOCK_VERSION 0
-- 
cgit v1.2.3-55-g7522


From 2b1c6bd77d4e6a727ffac8630cd154b2144b751a Mon Sep 17 00:00:00 2001
From: Christoph Hellwig
Date: Fri, 28 Nov 2008 10:09:09 +0100
Subject: generic compat_sys_ustat

Due to a different size of ino_t ustat needs a compat handler, but
currently only x86 and mips provide one.  Add a generic compat_sys_ustat
and switch all architectures over to it.  Instead of doing various
user copy hacks compat_sys_ustat just reimplements sys_ustat as
it's trivial.  This was suggested by Arnd Bergmann.

Found by Eric Sandeen when running xfstests/017 on ppc64, which causes
stack smashing warnings on RHEL/Fedora due to the too large amount of
data writen by the syscall.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 arch/ia64/ia32/ia32_entry.S        |  2 +-
 arch/mips/kernel/linux32.c         | 34 ----------------------------------
 arch/mips/kernel/scall64-n32.S     |  2 +-
 arch/mips/kernel/scall64-o32.S     |  2 +-
 arch/parisc/kernel/syscall_table.S |  2 +-
 arch/powerpc/include/asm/systbl.h  |  2 +-
 arch/s390/kernel/compat_wrapper.S  |  2 +-
 arch/x86/ia32/ia32entry.S          |  2 +-
 arch/x86/ia32/sys_ia32.c           | 22 ----------------------
 arch/x86/include/asm/ia32.h        |  7 -------
 arch/x86/include/asm/sys_ia32.h    |  2 --
 fs/compat.c                        | 28 ++++++++++++++++++++++++++++
 include/linux/compat.h             |  8 ++++++++
 13 files changed, 43 insertions(+), 72 deletions(-)

(limited to 'arch/s390')

diff --git a/arch/ia64/ia32/ia32_entry.S b/arch/ia64/ia32/ia32_entry.S
index a46f8395e9a5..af9405cd70e5 100644
--- a/arch/ia64/ia32/ia32_entry.S
+++ b/arch/ia64/ia32/ia32_entry.S
@@ -240,7 +240,7 @@ ia32_syscall_table:
 	data8 sys_ni_syscall
 	data8 sys_umask		  /* 60 */
 	data8 sys_chroot
-	data8 sys_ustat
+	data8 compat_sys_ustat
 	data8 sys_dup2
 	data8 sys_getppid
 	data8 sys_getpgrp	  /* 65 */
diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c
index 1a86f84fa947..784859cedef7 100644
--- a/arch/mips/kernel/linux32.c
+++ b/arch/mips/kernel/linux32.c
@@ -356,40 +356,6 @@ SYSCALL_DEFINE1(32_personality, unsigned long, personality)
 	return ret;
 }
 
-/* ustat compatibility */
-struct ustat32 {
-	compat_daddr_t	f_tfree;
-	compat_ino_t	f_tinode;
-	char		f_fname[6];
-	char		f_fpack[6];
-};
-
-extern asmlinkage long sys_ustat(dev_t dev, struct ustat __user * ubuf);
-
-SYSCALL_DEFINE2(32_ustat, dev_t, dev, struct ustat32 __user *, ubuf32)
-{
-	int err;
-	struct ustat tmp;
-	struct ustat32 tmp32;
-	mm_segment_t old_fs = get_fs();
-
-	set_fs(KERNEL_DS);
-	err = sys_ustat(dev, (struct ustat __user *)&tmp);
-	set_fs(old_fs);
-
-	if (err)
-		goto out;
-
-	memset(&tmp32, 0, sizeof(struct ustat32));
-	tmp32.f_tfree = tmp.f_tfree;
-	tmp32.f_tinode = tmp.f_tinode;
-
-	err = copy_to_user(ubuf32, &tmp32, sizeof(struct ustat32)) ? -EFAULT : 0;
-
-out:
-	return err;
-}
-
 SYSCALL_DEFINE4(32_sendfile, long, out_fd, long, in_fd,
 	compat_off_t __user *, offset, s32, count)
 {
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 7438e92f8a01..f61d6b0e5731 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -253,7 +253,7 @@ EXPORT(sysn32_call_table)
 	PTR	compat_sys_utime		/* 6130 */
 	PTR	sys_mknod
 	PTR	sys_32_personality
-	PTR	sys_32_ustat
+	PTR	compat_sys_ustat
 	PTR	compat_sys_statfs
 	PTR	compat_sys_fstatfs		/* 6135 */
 	PTR	sys_sysfs
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index b0fef4ff9827..60997f1f69d4 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -265,7 +265,7 @@ sys_call_table:
 	PTR	sys_olduname
 	PTR	sys_umask			/* 4060 */
 	PTR	sys_chroot
-	PTR	sys_32_ustat
+	PTR	compat_sys_ustat
 	PTR	sys_dup2
 	PTR	sys_getppid
 	PTR	sys_getpgrp			/* 4065 */
diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S
index 303d2b647e41..03b9a01bc16c 100644
--- a/arch/parisc/kernel/syscall_table.S
+++ b/arch/parisc/kernel/syscall_table.S
@@ -130,7 +130,7 @@
 	ENTRY_OURS(newuname)
 	ENTRY_SAME(umask)		/* 60 */
 	ENTRY_SAME(chroot)
-	ENTRY_SAME(ustat)
+	ENTRY_COMP(ustat)
 	ENTRY_SAME(dup2)
 	ENTRY_SAME(getppid)
 	ENTRY_SAME(getpgrp)		/* 65 */
diff --git a/arch/powerpc/include/asm/systbl.h b/arch/powerpc/include/asm/systbl.h
index 72353f6070a4..fe166491e9dc 100644
--- a/arch/powerpc/include/asm/systbl.h
+++ b/arch/powerpc/include/asm/systbl.h
@@ -65,7 +65,7 @@ SYSCALL(ni_syscall)
 SYSX(sys_ni_syscall,sys_olduname, sys_olduname)
 COMPAT_SYS_SPU(umask)
 SYSCALL_SPU(chroot)
-SYSCALL(ustat)
+COMPAT_SYS(ustat)
 SYSCALL_SPU(dup2)
 SYSCALL_SPU(getppid)
 SYSCALL_SPU(getpgrp)
diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index 62c706eb0de6..87cf5a79a351 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -252,7 +252,7 @@ sys32_chroot_wrapper:
 sys32_ustat_wrapper:
 	llgfr	%r2,%r2			# dev_t
 	llgtr	%r3,%r3			# struct ustat *
-	jg	sys_ustat
+	jg	compat_sys_ustat
 
 	.globl	sys32_dup2_wrapper
 sys32_dup2_wrapper:
diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index 5a0d76dc56a4..8ef8876666b2 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -557,7 +557,7 @@ ia32_sys_call_table:
 	.quad sys32_olduname
 	.quad sys_umask		/* 60 */
 	.quad sys_chroot
-	.quad sys32_ustat
+	.quad compat_sys_ustat
 	.quad sys_dup2
 	.quad sys_getppid
 	.quad sys_getpgrp		/* 65 */
diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c
index 6c0d7f6231af..efac92fd1efb 100644
--- a/arch/x86/ia32/sys_ia32.c
+++ b/arch/x86/ia32/sys_ia32.c
@@ -638,28 +638,6 @@ long sys32_uname(struct old_utsname __user *name)
 	return err ? -EFAULT : 0;
 }
 
-long sys32_ustat(unsigned dev, struct ustat32 __user *u32p)
-{
-	struct ustat u;
-	mm_segment_t seg;
-	int ret;
-
-	seg = get_fs();
-	set_fs(KERNEL_DS);
-	ret = sys_ustat(dev, (struct ustat __user *)&u);
-	set_fs(seg);
-	if (ret < 0)
-		return ret;
-
-	if (!access_ok(VERIFY_WRITE, u32p, sizeof(struct ustat32)) ||
-	    __put_user((__u32) u.f_tfree, &u32p->f_tfree) ||
-	    __put_user((__u32) u.f_tinode, &u32p->f_tfree) ||
-	    __copy_to_user(&u32p->f_fname, u.f_fname, sizeof(u.f_fname)) ||
-	    __copy_to_user(&u32p->f_fpack, u.f_fpack, sizeof(u.f_fpack)))
-		ret = -EFAULT;
-	return ret;
-}
-
 asmlinkage long sys32_execve(char __user *name, compat_uptr_t __user *argv,
 			     compat_uptr_t __user *envp, struct pt_regs *regs)
 {
diff --git a/arch/x86/include/asm/ia32.h b/arch/x86/include/asm/ia32.h
index 50ca486fd88c..1f7e62517284 100644
--- a/arch/x86/include/asm/ia32.h
+++ b/arch/x86/include/asm/ia32.h
@@ -129,13 +129,6 @@ typedef struct compat_siginfo {
 	} _sifields;
 } compat_siginfo_t;
 
-struct ustat32 {
-	__u32			f_tfree;
-	compat_ino_t		f_tinode;
-	char			f_fname[6];
-	char			f_fpack[6];
-};
-
 #define IA32_STACK_TOP IA32_PAGE_OFFSET
 
 #ifdef __KERNEL__
diff --git a/arch/x86/include/asm/sys_ia32.h b/arch/x86/include/asm/sys_ia32.h
index ffb08be2a530..72a6dcd1299b 100644
--- a/arch/x86/include/asm/sys_ia32.h
+++ b/arch/x86/include/asm/sys_ia32.h
@@ -70,8 +70,6 @@ struct old_utsname;
 asmlinkage long sys32_olduname(struct oldold_utsname __user *);
 long sys32_uname(struct old_utsname __user *);
 
-long sys32_ustat(unsigned, struct ustat32 __user *);
-
 asmlinkage long sys32_execve(char __user *, compat_uptr_t __user *,
 			     compat_uptr_t __user *, struct pt_regs *);
 asmlinkage long sys32_clone(unsigned int, unsigned int, struct pt_regs *);
diff --git a/fs/compat.c b/fs/compat.c
index d0145ca27572..4e0db94b5353 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -378,6 +378,34 @@ out:
 	return error;
 }
 
+/*
+ * This is a copy of sys_ustat, just dealing with a structure layout.
+ * Given how simple this syscall is that apporach is more maintainable
+ * than the various conversion hacks.
+ */
+asmlinkage long compat_sys_ustat(unsigned dev, struct compat_ustat __user *u)
+{
+	struct super_block *sb;
+	struct compat_ustat tmp;
+	struct kstatfs sbuf;
+	int err;
+
+	sb = user_get_super(new_decode_dev(dev));
+	if (!sb)
+		return -EINVAL;
+	err = vfs_statfs(sb->s_root, &sbuf);
+	drop_super(sb);
+	if (err)
+		return err;
+
+	memset(&tmp, 0, sizeof(struct compat_ustat));
+	tmp.f_tfree = sbuf.f_bfree;
+	tmp.f_tinode = sbuf.f_ffree;
+	if (copy_to_user(u, &tmp, sizeof(struct compat_ustat)))
+		return -EFAULT;
+	return 0;
+}
+
 static int get_compat_flock(struct flock *kfl, struct compat_flock __user *ufl)
 {
 	if (!access_ok(VERIFY_READ, ufl, sizeof(*ufl)) ||
diff --git a/include/linux/compat.h b/include/linux/compat.h
index 3fd2194ff573..b880864672de 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -125,6 +125,13 @@ struct compat_dirent {
 	char		d_name[256];
 };
 
+struct compat_ustat {
+	compat_daddr_t		f_tfree;
+	compat_ino_t		f_tinode;
+	char			f_fname[6];
+	char			f_fpack[6];
+};
+
 typedef union compat_sigval {
 	compat_int_t	sival_int;
 	compat_uptr_t	sival_ptr;
@@ -178,6 +185,7 @@ long compat_sys_semtimedop(int semid, struct sembuf __user *tsems,
 		unsigned nsems, const struct compat_timespec __user *timeout);
 asmlinkage long compat_sys_keyctl(u32 option,
 			      u32 arg2, u32 arg3, u32 arg4, u32 arg5);
+asmlinkage long compat_sys_ustat(unsigned dev, struct compat_ustat __user *u32);
 
 asmlinkage ssize_t compat_sys_readv(unsigned long fd,
 		const struct compat_iovec __user *vec, unsigned long vlen);
-- 
cgit v1.2.3-55-g7522